Session Recap: Efficacy of College Affirmative Action and Alternative Policies
November 9, 2014 09:00 AM
By Alix Gould-Werth, University of Michigan
In the past decade, a series of judicial decisions have constrained the ability of public colleges and universities to use race as an explicit factor in admissions decisions. The papers on this panel used empirical data and simulations to examine how university admissions practices—changed in the wake of these judicial decisions—affect the composition of the student body, both in terms of racial composition and levels of student academic preparedness.
Simulation Models of the Effects of Race- and Socioeconomic-Based Affirmative Action Policies by Sean F. Reardon, Rachel Baker, and Matt Kasman, Stanford University; Daniel Klasik, University of Maryland; Joe Townsend, Stanford University
The authors used agent-based modeling to investigate how various affirmative action policies affect the demography composition of schools, particularly examining the efficacy of using socioeconomic status as a proxy for race. The main takeaway from this paper was that affirmative action on the basis of socioeconomic status does not diversity the student body racially to the extent that affirmative action based on race would: in the pool of college applicants race and socioeconomic status are not sufficiently correlated for socioeconomic status to serve as a proxy for race.
Is There a "Workable" Race-Neutral Alternative to Affirmative Action in College Admissions? by Mark C. Long, University of Washington
Long used simulation to contrast the composition of student bodies under three affirmative action regimes: traditional race-based affirmative action, a passive affirmative action ban, and proxy based affirmative action. He asked whether proxy-based affirmative action can be used to generate the same share of racial minority students as under traditional affirmative action, while maintaining level of academic preparedness. The simulations showed that using proxy-based affirmative action is associated with a fall in average GPA in comparison to race-based affirmative actions, and consequently a predicted drop in rates of graduation—by one percentage point.
Affirmative Action and University Fit: Evidence from Proposition 209 by Peter Arcidiacono, Duke University; Esteban M. Aucejo, London School of Economics; Patrick Coate, University of Michigan; V. Joseph Hotz, Duke University
This paper turns to observed outcomes, using administrative data to investigate how Proposition 209, which banned affirmative action in California, affected racial minority enrollment and graduation rates in the UC system. The paper deals with the empirical finding that graduation rates increased following the affirmative action ban, and asks whether this change can be linked to changed admissions processes. This paper argues that students were more efficiently matched to institutions following the cessation of affirmative action, and that institutional practices also evolved to better support enrolled students in the wake of Proposition 209.
Comments from discussants and audience members pushed the panelists in two major directions. First, they asked panelists to engage directly with the policy motivations and implications of their work: why is diversity valued on college campuses? What implications do their results have for college admissions practices? Second, panelists were asked to engage with the institutional context in which they observed (or simulated) responses to admissions practices: the dependent variable of interest is academic success in the college context, but when admissions policies change the composition of the student body, universities change supports for these students. Panelists were asked to acknowledge, and perhaps, model institutional practices that co-vary with changes to admissions practices and also affect student success.