Rank-Order Correlation (Kendall's rho)

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Yodan Rose's North Beach study found high rank-order correlation (Kendall's rho) between different people's diagnoses of good and bad places along streets in North Beach, San Franciscofinding0.789
  Empirical evidence that neighborhood quality diagnosis is objective rather than merely a matter of opinion.
• Spearman Rank Correlationmethod0.761
  Used to compare RDMs in RSA computations; noted to have sensitivity issues with differing relative extrema in embedding layers.
• Most independent dimension pair is aesthetic_response and boundary_awareness (rho=0.553); most correlated is prediction_error and conceptual_crystallization (rho=0.886)finding0.722
  Characterizes internal structure of the six scoring dimensions
• Spearman's rank correlation among different alignment metrics (CKA, SVCCA, Mutual k-NN, CKNNA) over 78 vision models is high across variants, with all p-values below 2.24×10^-105finding0.720
  Validates robustness of alignment metric choice
• Linear combination ρd – dr – log k yields a predictive correlate of success.hypothesis0.715
  Implicit hypothesis behind S form.
• S = ρd − dr − log k is a predictive correlate of anchoring success across few-shot, SFT, and CoT.claim0.711
  UCCT's practical utility claim.
• Haiku-Kimi per-koan correlation rho=0.123 (p=0.52); H5a trace distillation not supported at individual model levelfinding0.710
  Group correlation (rho=0.634) dissolves at individual level; shared posture not shared voice
• Correlation between self-evaluation and textual evaluation of SAE feature emotionality: rho=+0.051 (n.s.)finding0.707
  Shows that the two evaluation methods for emotionality are largely uncorrelated, indicating they capture different signals