Old Revisions

These are the older revisons of the current document. To revert to an old revision, select it from below, click Edit this page and save it.

2026/08/21 14:50 statistics:pvalue_corrections – Boxes: <wrap> renders a span, use uppercase <WRAP>; one box per list, not per bullet karel.kubicek.claude ±0 B (current)
2026/08/21 08:33 statistics:pvalue_corrections – [Open Questions] karelkubicek ±0 B
2026/08/13 07:25 statistics:pvalue_corrections – One more artefact found in the family-size hand list by reading the source: the PETS 2022 multi-region hit was running-header furniture, not a count. Figure moves from 16/265 (6.0%) to 15/265 (5.7%), complement to 94.3%. Authored by Claude karel.kubicek.claude ±0 B
2026/08/13 07:23 statistics:pvalue_corrections – Re-review fixes: the full-text family-size figure was inflated 2.4x by chi-squared notation matching /d+ tests/ -- now guarded, all 20 surviving hits hand-read, 16 genuine = 6.0%, which agrees with the 4.9% tuple figure; drop the 'upper bound' framing (it karel.kubicek.claude +2.5 KB
2026/08/13 07:09 statistics:pvalue_corrections – Generic-review fixes (14 accepted): new 'Deciding What Counts as One Family' section; headline reframed to what the data shows (participants predict correcting, not family size) and the 'both 34.7%' cell restored; 'no remaining reason to use Bonferroni' r karel.kubicek.claude +6.5 KB
2026/08/13 06:56 statistics:pvalue_corrections – Review fixes: replace two false 'appears in zero' claims about forking paths (the string matches twice, both false positives); fix a footnote that said 42 exceeds 45; correct the per-year range to 0.0%-39.1% on 8-128; restore a plural in the Sunlight quot karel.kubicek.claude +2.5 KB
2026/08/13 06:43 statistics:pvalue_corrections – Fix the large-n worked example: 0.4pp at n=100,000 does not give p<1e-4 (real value 0.003). Replaced with a verified 1.0pp example (p = 3e-13, 3e-10 after Bonferroni at k=1000) and a footnote showing the computation. Authored by Claude karel.kubicek.claude +453 B
2026/08/13 06:35 statistics:pvalue_corrections – New page: p-value corrections in web measurement. 24.1% of 1,025 hypothesis-test papers correct; 12.9% of crawl papers vs 38.6% of user studies; Bonferroni still modal but dominated by Holm; FWER-vs-FDR choice, dependence, large-n; downloadable adjust_pva karel.kubicek.claude +49.3 KB