Perceptions Of Required Ability Act As A Proxy For Actual Required Ability In Explaining The Gender Gap

A rigorous takedown of Leslie et al. (2015), which claimed women are underrepresented in fields believed to require innate "brilliance" because women are stereotyped as lacking it (perceived-ability r = -0.60 with % women). Scott shows the perception is just a proxy for actual ability: GRE-Quantitative scores correlate with % women at r = -0.82 (stronger), the perceived-ability effect is statistically mediated by actual ability (like "perception of smoking" predicting cancer only via real smoking), and a simple GRE-threshold model predicts the math gender ratio almost exactly. He pre-empts the STEM-binary, undergrad-precursor (the SAT gap is already there by 11th grade), stereotype-threat, and reverse-causation objections.

Quality 79 · Excellent. A careful, statistically sharp, brave debunk-plus-alternative-model; the proxy/mediation structure and the GRE-threshold prediction are genuinely illuminating. Held below the top by being a single-study rebuttal on a narrow (if charged) question. Excellent, lower end.

Claude’s paradigm shift 35 · Slight. Applies a proxy/mediation argument and threshold reasoning rather than introducing a new concept; sharp but not a new paradigm. Slight, edging moderate.

Real-world impact 2 · Minor. A careful, statistically sharp, brave debunk-plus-alternative-model (perceived 'brilliance' is just a proxy for actual GRE-Q ability) on a charged question. Methodological influence, but a single-study rebuttal of narrow reach with no material change — low RWI.