Background
This is level 4 of 5 and continues the work of the earlier levels. The new files for this level are in workspace/t4/. The records from the start are in workspace/t1/, and the earlier results are in workspace/t1/results/ to workspace/t3/results/.
Task
Hi,
Imani and I have added the specialized-rater-pool releases for the annotation robustness section. We need the supplementary rater comparisons, and a cross-campaign cohort for the model section that will hold up.
The panel workbook goes in workspace/t4/results/. Please quantify the paired differences between pools, with simultaneous uncertainty across comments and raters for the pool and contributor-treatment comparisons, and show how far the unresolved responses could move those differences. Unsure and incomprehensible responses should be counted separately.
Use the earlier corpus and context work to link the annotation campaigns. Keep the evidence behind each linkage choice, and the alternatives that could change the cohort or the annotations attached to it. We'll run the model reproduction on these cohorts, so we need to be able to tell a change of linkage apart from a change of annotation reference. When you compare the linked and complete panels, use the same finite-panel treatment as in the replication.
Best, Noor
Input
workspace/t1/: the records supplied with the work (downloads/,research/,tooling/).workspace/t2/: the files that came with level t2 (context_toxicity/).workspace/t3/: the files that came with level t3 (context_toxicity/,wiki-detox/).workspace/t4/: the files that came with level t4 (jigsaw-specialized-rater-pools-dataset/,research/).workspace/t1/results/: the results of level t1.workspace/t2/results/: the results of level t2.workspace/t3/results/: the results of level t3.
Output
workspace/t4/results/rater_panel.xlsx