SAFS Logo

Research war on affirmative action

September 2012

With the U.S. Supreme Court about to hear arguments in a case that could decide the fate of affirmative action in admissions, a research war has broken out. Defenders and critics of the consideration of race are releasing new studies (some of which were submitted in briefs to the court) on the impact of affirmative action.

Several studies presented Friday at the Brookings Institution suggested that eliminating the consideration of race would not have as dramatic an effect on minority students as some believe, and that the beneficiaries of affirmative action may in fact achieve less academic success than they would otherwise. The studies were criticized by some present for being one-sided.

Yield Rates After a Ban

One of the studies explored the impact on the yield rate – the proportion of admitted students who enroll – of black, Latino and Native American students to the University of California after the state barred the consideration of race in admissions. Minority enrollments fell after voters adopted the ban, in part because greater proportions of white and Asian applicants, on average, have the academic and other credentials required for admission.

But the paper focused on a subset of minority students: those who gained admission after the ban on affirmative action. Many observers at the time predicted that these students would be less likely to enroll, as they would perceive the system to be hostile, and news articles quoted minority students to this effect.

But a paper by Kate Antonovics, an economist at the University of California at San Diego, and Richard Sander, a law professor at the University of California at Los Angeles, found that the ban on affirmative action didn't produce "chilling effects," but actually produced "warming effects" on the likelihood of minority students' accepting offers of admission.

The paper examined the yield rates by students in the three years after Proposition 209 (the ban) took effect. Across the system, the yield rate increased by about 10 percent. The increase was the greatest at Berkeley and UCLA. At Berkeley, the pre-Proposition 209 yield rate for underrepresented minority groups was 37.9 percent, and that increased by 5.7 percentage points after affirmative action considerations were barred from the admissions process. At UCLA, a 38.8 percent yield rate increased by 3.9 percentage points.

Antonovics, who presented the paper, said that the findings were important. She said that critics of affirmative action might expect (and accept) that the end of the consideration of race would lead to an overall decline in minority enrollment rates, but that there would be "a layer of concern" if some of that drop could be attributed to students who met race-neutral admissions standards opting not to enroll. She said that she and Sander expected to find this "chilling effect," and were surprised to find "no evidence for it" and instead see evidence showing the opposite impact.

'Mismatch' Theory

The fear Antonovics expressed is that minority students might not be enrolling at highly competitive institutions at which they would be a good academic match. Another kind of "mismatch" has also been the focus of much research on affirmative action in recent years -- and that idea (hotly debated) is that minority students who are admitted to better institutions because of affirmative action may end up with lower academic achievement as a result. (Sander, who helped organize the Brookings program, has been a leading proponent of the idea.)

On Friday, Doug Williams, chair of economics at the University of the South, presented a paper arguing that law schools provide solid evidence for the existence of mismatch. He said that because of the requirement that lawyers pass a bar exam, there is an independent measure of successful learning in law school. Because some minority students with similar academic credentials (judged by undergraduate grads and LSAT scores) enroll at more or less competitive law schools, he said it is possible to see whether there is a mismatch.

There has long been a gap between black and white law graduates on bar passage rates, he noted. Currently, about 97 percent of white law school graduates pass the bar, with 92 percent passing on the first try. For black law graduates, the figures are 78 and 61 percent, respectively. But Williams said that these gaps are partly explained by the differing academic credentials of black and white law students, and so by themselves don't back mismatch theory.

But in comparing minority law graduates of comparable academic preparation who enroll in more or less competitive law schools, Williams said that those (from among those with comparable academic credentials) who enroll in the bottom two tiers of law schools pass the bar at a rate that is 14 percentage points higher than that of those who enroll at the top law schools. And he said that suggested a statistically significant mismatch.

Williams said that he would like more detailed information about law students – including the law schools that they apply to and are admitted to – but based on the information he does have, he said that he had "strong evidence" to back mismatch.

A One-Sided Presentation?

In the question period after those two presentations, Richard O. Lempert, an emeritus professor of law and sociology at the University of Michigan, criticized not only those papers, but the entire way the program at Brooking was organized. "All of the papers are on one side of the debate," Lempert said. He questioned why "first rate" scholars who support affirmative action were not presenting work, and said that the papers presented had "serious methodological" issues. He also said that the papers were not peer-reviewed. Antonovics responded by saying that she had mixed feelings about affirmative action and did not do this research with any outcome in mind. Further, she said that the paper she presented has just been accepted for publication by a peer-reviewed economics journal.

In terms of the substance of the papers, Lempert said that a greater chilling effect would be seen on applications than in yield. (Antonovics said that application volume did not change much after Proposition 209.) And Lempert said that the analysis by Williams ignored the way most black law students cluster in the middle tiers of law schools, not the top or bottom tiers, and he said that one key factor in the bottom tiers was that they included historically black colleges.

Williams acknowledged that the success of black students at historically black law schools could be attributed to a number of factors. But he said it would be wrong to exclude those institutions from the study.

Sander said that panelists at the program were "ideologically diverse" and that there was no attempt to include only scholars of one point of view.

The debate will likely continue Thursday when a coalition of social science groups -- the American Educational Research Association, the American Sociological Association, the American Statistical Association, the Association for the Study of Higher Education, the Law and Society Association and the Linguistic Society of America -- hold a briefing on research those organizations cited in their brief to the Supreme Court encouraging the justices to uphold the legality of the consideration of race and ethnicity in admissions.