August 14, 2012 — On Monday, August 13, 2012, a group of leading empirical scholars with over 255 combined years of social-science research experience filed a brief amici curiae in support of respondent University of Texas in Fisher v. University of Texas at Austin, the U.S. Supreme Court case concerning the constitutionality of The University of Texas’ affirmative-action admissions policy. The brief was authored by Thomas S. Leatherbury, Harry M. Reasoner, Kimberly R. McCoy, and Eric A. White of Vinson & Elkins LLP on behalf of Professors Guido Imbens (Stanford), Donald B. Rubin (Harvard), Gary King (Harvard), Richard A. Berk (Univ. of Pennsylvania), Daniel E. Ho (Stanford), Kevin M. Quinn (UC Berkeley – Boalt Hall), D. James Greiner (Harvard), Ian Ayres (Yale), Richard Brooks (Yale), Paul Oyer (Stanford), and Richard Lempert (Univ. of Michigan).
Their brief points out the “substantial methodological flaws” in the amicus briefs relying on the “mismatch” hypothesis propounded primarily in research by Richard Sander, a UCLA professor, and Doug Williams of the University of the South (Sewanee). The empirical scholars’ brief conclude that the mismatch research “fails to satisfy the basic standards of good empirical social-science research.”
Relying upon Sander and Williams, other amici argue to the Court that social science has shown that affirmative action is harmful to minority students. According to the “mismatch” hypothesis, when a minority student attends a more selective college or graduate school as a result of race-based admissions, and his academic credentials are substantially below those of his classmates, he does not thrive. For example, Sander and Williams argue, as a result of being surrounded by students with stronger academic credentials in a higher-ranked law school, a minority law student may end up at the bottom of the class and end up failing the bar exam.
Citing over a dozen studies, the empirical scholars show that mismatch research by Sander and Williams and its conclusions have been subject to wide-ranging criticism on methodological grounds and does not represent a consensus in social science. That research purports to draw an inference, for example, about the causal effect of attending a higher law school tier. The brief articulates how this research fails to abide by widely accepted principles of causal inference.
- In order to draw a causal inference, researchers should generate (a) comparison groups that are (b) as similar as possible in pre-existing characteristics, so that (c) differences in outcomes can be attributed to the selectivity of the institution.
- Sander’s analyses draw invalid comparisons between black students at selective institutions and white students at less-selective institutions. The analyses thereby fail to hold constant pre-existing attributes other than law school tier.
- Sander and Williams’ research adjusts not only for pre-existing characteristics, but also for outcomes that law school tier — according to the mismatch hypothesis — affects. Adjusting for outcomes, such as law school graduation or law school grades, is inconsistent with the mismatch hypothesis and contaminates inferences about the causal effect of law-school tier.
- Analyses of the same data that hold constant pre-existing characteristics (undergraduate GPA scores, LSAT scores, race, and gender) have not found any substantially and statistically significant effects of law-school tier on bar passage.
- The amici provide examples of better-designed studies that apply principles of causal inference. Stacy B. Dale, a Senior Researcher at Mathematica Policy Research, and Alan B. Krueger, the Bendheim Professor of Economics and Public Affairs at Princeton University, for example, generally found no statistically significant effect of attending a selective university on earnings. The exception was for black, Latino, and low-income background students who — contrary to the mismatch hypothesis — experience a significant increase in earnings.
- The amici conclude the mismatch research should not be relied upon by the Court in reaching a decision in the case.
Vinson & Elkins LLP is an international law firm with more than 700 lawyers across 16 offices worldwide. For further information, please contact Daniel E. Ho at firstname.lastname@example.org or (650) 723-9560 or Tom Leatherbury at email@example.com or (214) 220-7792.