Resampling cuts point estimates ~12% and inflates standard errors ~40%, replication thread shows
RexDouglass · x · 2026-09-04
- A post from the I4Replication thread quantifying how sampling variation destabilizes empirical results: after resampling, point estimates fall 12% while standard errors rise 40%.
- Part of a broader argument that data sampling noise alone is a major contributor to the fragility of published empirical findings.
Related event: Resampling study shakes significance of top-journal findings(3 posts)→
More from Research
- Last Translation Benchmark Is Live: Contribute 10 Examples, Get Co-authorship — zouharvi · 2026-09-04
- Translation Models Fail Beyond Figurative Language, Says Benchmark Author — zouharvi · 2026-09-04
- Last Translation Benchmark: 3456 Crowdsourced Examples Break SOTA Translation Models — zouharvi · 2026-09-04
- Machine Translation Is Not Solved: Last Translation Benchmark Released — zouharvi · 2026-09-04
- Benchmarks Saturate Faster Than Ever, But New Ones Lack Validation, Says Author — subinium · 2026-09-04
- Is recurrent depth the real driver? Researchers debate 3x depth vs 81x params math — xuanalogue · 2026-09-04