Loading...
In the fall of 1973, the University of California, Berkeley found itself staring at damning numbers. Of 8,442 men who applied to its graduate programs, 44% were admitted. Of 4,321 women who applied, only 35% were accepted. The gap was stark enough that the university feared a federal sex discrimination lawsuit. Statistician Peter Bickel was brought in to investigate. What Bickel discovered—and published in Science in 1975 alongside co-authors E.A. Hammel and J.W. O'Connell—upended everything. When he broke down the admissions data department by department, the picture reversed. In most of Berkeley's 85 departments, women were admitted at equal or slightly higher rates than men. In the English department, women outpaced men. In some STEM fields, the female admission rate exceeded the mal...
Popular framing: Berkeley discriminated against women in graduate admissions.
Structural analysis: The aggregate 35-vs-44% gap was real, but the causal mechanism wasn't at the decision node — women applied disproportionately to oversubscribed, low-admit-rate departments (English, Education) while men clustered in high-admit-rate STEM fields. Simpson's paradox surfaces because the choice of aggregation embeds a causal assumption; base-rate neglect at the headline level converted a path-dependent funneling problem upstream (which departments women believed they belonged in) into a discrimination charge at the wrong layer. The deeper structural force was societal — not the admissions office.
The popular framing conflates procedural fairness at one stage with systemic equity across the whole pipeline, committing base-rate neglect at a social level — treating 'department choice' as a neutral variable when it is itself a downstream effect of prior discrimination. Recognizing this gap matters because institutions can declare victory on the measurable proximate cause while the structural roots remain untouched and politically invisible.
