Wilcoxon Signed Rank Test Beats McNemar for Binary Evals
Follow-up validation of evalstats finds the Wilcoxon signed rank test outperforms the commonly recommended McNemar mid-p test for paired p-value tests on binary evaluation data.
2026-08-31 ~ 2026-08-31 · 2 related posts
- Wilcoxon signed rank test outperforms McNemar for binary data evals — IanArawjo · 2026-08-31
- Wilcoxon Signed Rank Outperforms McNemar for Binary Eval Data — IanArawjo · 2026-08-31