Following Justice Scalia's controversial question about mismatch research at the Fisher v. Texas oral argument, some commentators have argued – sometimes civilly, sometimes less so – that whatever the merits of the mismatch hypothesis, it is merely a policy question irrelevant to the core constitutional question at stake. At the Chicago-Kent Law School’s iSCOTUS blog, for example, professor Vinay Harpalani writes “'mismatch’ is irrelevant to the constitutional debate on race-conscious admissions. The Supreme Court is not an admissions committee: universities possess much greater expertise than courts on determining the academic qualifications of students.” Elsewhere, Paul Campos at Salon writes that “Something that has been obscured by the outcry over Scalia’s rhetoric is that this argument – that affirmative action is bad for its putative beneficiaries because it pushes them into settings where they will be less likely to excel – is, from a legal standpoint, completely irrelevant to the issue before the Supreme Court. That is because the legal issue in the case is whether affirmative action programs such as that being employed by the University of Texas are unconstitutional, not whether they are a good idea.”

The mismatch critique of affirmative action certainly does present important policy questions. I’m glad that Scalia’s question, though less than ideally phrased, is getting people to talk about them. But with all due respect to Harpalani and Campos, mismatch is a constitutional issue too.

To understand why, let us review for a moment the relevant constitutional doctrine as developed by Grutter and Fisher I. Grutter holds that to be constitutional, a race-preferential admissions program is constitutional if it is narrowly tailored to serve a university’s a compelling interest in obtaining the educational benefits that flow from having a diverse student body. Narrow tailoring further requires that a race-conscious admissions program not unduly harm members of any racial group (Grutter at 341). Fisher I clarified that, while a university’s educational judgment is due some deference regarding the compelling interest part of the test, courts do not defer to the university when analyzing narrow tailoring.  

The mismatch research indicates that, because of large racial preferences, some racial and ethnic minority students learn less than they would in settings where their entering academic credentials are close to those of the median student. Or, phrased slightly differently, large preferences put minority students in a worse educational and pedagogical position than they would be if such preferences didn’t exist. 

So the relevant questions are: can a race-preferential admissions program really be narrowly tailored to achieve the educational benefits of diversity for all students if it makes a significant subgroup of students educationally worse off than they would be under race-neutral admissions? Is the evidence of mismatch at UT strong enough that students of some racial groups are unduly harmed? What if it can be shown that UT didn’t even bother to look at the mismatch research when formulating its admissions policies concerning race? Shouldn’t a university carefully calibrating its program to use race no more than necessary to achieve diversity’s educational benefits be paying close attention to research showing preferences’ possible downsides? 

Also, if a given program routinely leads students to be mismatched academically, is it really achieving the educational benefits of diversity? Do academically mismatched students become less confident about speaking up in class to share their distinct experiences or perspectives? Will they hesitate to participate in dorm room bull sessions because they feel outgunned? Rick Sander points to the phenomenon of "social mismatch" -- students tend to socialize mostly with those who have similar academic credentials and performance. Very large preferences may therefore inadvertently contribute to racial isolation in social settings. Can admissions practices that do this really be considered to further a compelling state interest in diversity?

The “narrow tailoring” part of the Grutter/Fisher I test exists in part to identify admissions programs that use race more than is necessary to achieve the educational benefits of diversity. But it also helps courts smoke out race-preferential admissions programs that are not really educational diversity programs, but that instead promote “social justice,” proportional racial representation for its own sake, or some other constitutionally impermissible goal (See Grutter at 333, quoting Richmond v. J. A. Croson Co., 488 U. S., at 493 (plurality opinion) (“the means chosen [must] `fit' th[e] compelling goal so closely that there is little or no possibility that the motive for the classification was illegitimate’”). It makes little sense for a university to ignore the mismatch research if its reasons for using race-preferential admissions are educational or pedagogical. It makes more sense to ignore mismatch if the university is instead pursuing other goals. 

Yes, answering the questions that I pose above will require courts to engage with complex and occasionally difficult to understand social science research, a role that many Federalist Society types (including me) worry that judges are ill-equipped to play. Campos’s Salon piece on Fisher is right to note the problems that commonly accompany judicial forays into social science. All of this is one more reason why the Supreme Court should rethink the core doctrine of Grutter. So long as Grutter and Fisher I remain the law of the land, however, judges not only can, but must, consider the mismatch research.