Affirmative action still hasn’t been shown to reduce the number of black lawyers: A response to Sander

Abstract

Sander (2019) attempts to revive the claim that “mismatch” between the credentials of students that receive racial preferences in law school admissions and the average observable academic credentials of their peers leads to fewer black lawyers. This article examines Sander’s study and explains the reasons why second-choice analyses, and Bar Passage Study data in particular, are poor sources for causal inferences about academic mismatch. Sander’s paper makes indefensibly strong assumptions about the inferences that can be drawn between evidence on distinct types of mismatch, overclaims results that lack robustness across different subsamples of the underlying data, and misinterprets other results that in fact cut against the article’s claim. Ultimately, as originally reported in Ayres & Brooks (2005), the data do not provide evidence that affirmative action reduces the number of black lawyers.