rsa-exception-to-width-bias

IN premise — summaries/2026/08/24/aristotelian-2026-sR-references-chunk-2.md

Created 2026-08-24T17:10:51+00:00

RSA (Spearman rank correlation of dissimilarity matrices) is an exception to width-driven bias: as a self-normalized correlation of dissimilarities, its null stays near zero with no width-driven drift even without calibration

Summary

RSA avoids the common problem where a test's baseline drifts away from zero simply because you added more items or dimensions to your analysis. Because it is built from a self-normalizing rank correlation, the result stays centered near its expected value on its own, so you don't need extra corrections to keep your comparisons honest as the problem grows wider.