Student Research

Do Affirmative Action Bans Discourage Minority Applicants from Applying to College?

By Ben Backes

Affirmative Action in College Admissions

Affirmative action in college admissions continues to be a central issue in American higher education. Sometime in 2012, the Supreme Court will decide whether to hear a case challenging the use of racial preferences by the University of Texas. Although the debate is often focused on abstract principles — if past inequities should be corrected by differential treatment — several empirical questions are prominent in the discussion.

For example, proponents of affirmative action often argue that the abolition of racial preferences sends a message of institutional hostility to potential underrepresented minority students, discouraging applications and further lowering the already depressed college graduation rates of blacks and Hispanics. Indeed, the University of Texas successfully defended its re-introduction of racial preferences in a 2011 decision from the 5th U.S. Circuit Court of Appeals by arguing that without racial preferences, highly qualified minority students would be less likely to apply for admission.

In a recent paper co-authored with Kate Antonovics, we set out to explore whether California’s 1998 ban on affirmative action — voted into law as Proposition 209 — reduced the number of applications from minority students to UC campuses. In the paper, we look at admissions and application rates and test for evidence of a change in application behavior not directly explained by falling admissions chances.

The Picture at Berkeley

As a starting point, we examine admission and application rates at UC Berkeley. Since both of these rates vary greatly by academic credentials, we use an index of GPA (40 percent of the total weight) and SAT scores (30 percent each for math and verbal) to characterize the response to Prop 209.

Figure A shows how the removal of racial preferences affected the probability of admission for the entire range of academic credentials. The x-axis measures academic credentials (with the most highly qualified applicants on the right), and the y-axis shows average admissions rates. Thus, for underrepresented minority (URM) students and non-URM students with a given level of academic credentials, the figure shows average admissions rates for each group. As expected, applicants with higher academic credentials are more likely to gain admission. Additionally the figure shows that prior to Prop 209, there was a large disparity between the admissions rates of URM (top red line) and white (top green line) applicants. (Using Asians — rather than whites — as a control group gives similar results.) However, after Prop 209, most of the advantage was removed: The red and green lines in the post period are much closer together.

Figure B shows the corresponding graph for application rates. Despite the large drop in admissions probabilities shown in Figure A, minority students remained much more likely to apply to Berkeley than whites with similar academic credentials. It does not appear that the large drop in admissions probability led to a reduction in application rates.

Remaining UC Campuses

For the remainder of the UC campuses, we have two main findings.

First, all URM students, regardless of their academic credentials, experienced a substantial relative drop in their chances of admission to at least one UC campus. Even the most highly qualified applicants had significant drops in admissions probability to Berkeley and UCLA, while students with low GPA and SAT scores were less likely to get into less selective UC campuses, such as Santa Cruz and Riverside.

Second, the relative decline in URM application rates after the end of affirmative action was concentrated at Berkeley and UCLA, where students experienced the largest drops in their predicted probability of admission. The fall in application rates, however, was small relative to the drop in probability of admission. In addition, these URM students increased the number of applications they sent to less selective UC campuses. This had the effect of lowering the average quality of the set of UC campuses to which URM students applied, though the magnitude of this drop was small.

Discussion

An important issue in the debate surrounding Prop 209 — and bans on affirmative action more generally — is whether the end of the use of racial preferences lowered the value URM students placed on attending UC campuses. For example, URM students may have become less interested in applying to UC schools after the ban because they feared, justifiably, that they would have fewer same-race peers if they were admitted. Taken at face value, our results are not broadly consistent with the idea that Prop 209 dramatically lowered the interest of URM students in attending UC campuses.

In evaluating why the total enrollment numbers of URM students fell dramatically at schools like Berkeley and UCLA after the end of racial preferences, the results of our paper suggest that this fall was brought about by the direct effect of the drop in URM students’ chances of admission and not because URM students were discouraged from applying. URM-student interest in the UC system appears to have remained high. URM students with a given level of GPA and SAT scores were more likely than whites to apply, and be admitted, to UC campuses both before and after the ban on affirmative action, especially at the most selective campuses. This suggests that efforts to increase diversity in the UC system should be targeted at increasing URM students’ level of academic preparation before they apply for college rather than at increasing application rates.