Does Affirmative Action Create Mismatches Between Students and Universities?

The debate over ‘mismatch theory’ is needlessly polarized—and further research will help students, whatever the answers it generates.

Darren Ornitz / Reuters

Last week, during oral arguments in the Fisher v. University of Texas affirmative action case, Supreme Court Justice Antonin Scalia drew outraged criticism for declaring that “there are those who contend that it does not benefit African­-Americans to get them into the University of Texas where they do not do well, as opposed to having them go to a less­ advanced school, a slower­-track school where they do well.” Scalia was clumsily alluding to “mismatch theory,” a prominent critique of affirmative action. Its proponents argue that non-academic preferences in college admissions ill-serve some intended beneficiaries, who end up admitted to schools for which they are relatively unprepared, and struggling, rather than thriving at different schools where they would be at least as well prepared as their classmates.

The denunciations were fierce.

“It is deeply disturbing to hear a Supreme Court justice endorse racist ideas from the bench on the nation's highest court,” said Senate Minority Leader Harry Reid. The cover of the New York Daily News declared it “Scalia’s racist rant.” Andy Borowitz, whose humor reliably flatters the preexisting beliefs of the average New Yorker reader, quipped that Justice Antonin Scalia would fare better if he served as a judge at a court that was ‘less advanced’ than the United States Supreme Court.”

In response, some indignant conservative media critics declared that “mismatch theory” is “a taboo subject for the MSM” and that “the left” is made up of “contemptible liars” who are “terming Scalia a racist” because “the left’s ugly belief that subjective hurt somehow trumps statistical fact means that he must be labeled a bigot.”

If the underlying subject were inconsequential, America’s red and blue tribes could watch the uncharitable back-and-forth, get their respective serotonin boosts, and depart with the satisfaction of feeling superior to their ideological adversaries. As is often the case, however, the hyperbolic sniping that emerged from the political culture of stigma-and-outrage junkies muddied an important debate about a subject that deserves to be engaged cooly, rigorously, and constructively.

After all, many institutions of higher education give admissions boosts for non-academic characteristics, including race, athletic ability, musical talent, leadership potential, geographic diversity, and having parents who are alumni. It would be beneficial for the relevant admissions officers to have empirical answers to questions like, “Does an admitted student’s graduation rate change predictably depending on how many standard deviations they are from the mean SAT score or GPA?”

Colleges can hardly avoid provisional, working answers to that question. And there are cases in which nearly everyone either agrees with, or rejects, “mismatch theory.” If Harvard’s admissions team unexpectedly decided to admit ten students, never mind their race, who scored in the bottom 10 percent on the SAT, everyone would expect those students to fail. If Duke admitted 10 students of any race who scored a single percentile lower than the school average on the SAT, no one would expect that tranche of students to fail out of the university at a higher rate. Admissions officers regularly say about people of all races, “I’m impressed by this young person, but their ACT score and GPA make me wonder if they’d thrive here or be set up to fail.” Treating that commonplace as taboo is irresponsible. The only question is at what point academic credentials matter and to what degree.

With better information, colleges might learn that more expansive race-based preferences would not lower graduation rates; or that a subset of orchestra members or legacies or minorities are harmed by policies intended to help them; or that institutions can eliminate differences in graduation rates among students with GPA disparities by investing in specific types of academic support. Accurate, detailed conclusions could plausibly improve many thousands of lives, whatever they say.

Yet that isn’t the focus of the public debate. Why?

Scalia’s error was to talk carelessly and imprecisely about a predictably fraught subject. Contrary to his lazy characterization, proponents of “mismatch theory” do not believe that admission to selective colleges “does not benefit African ­Americans,” full stop, or that African Americans would benefit from “a slower track school.”

He ought to have made all of the following clear:

  • Many black students are fully qualified to attend the most highly selective institutions of higher education in America, and proponents of “mismatch theory” of course believe that this subset of black students benefits from doing so.
  • Professor Richard Sander of UCLA, who many regard as the foremost scholarly proponent of “mismatch theory,”and Stuart Taylor Jr. of Brookings, who co-authored the book Mismatch with him, “support the modest use of race in admissions but think very large preferences have harmful effects.”
  • It’s not about just about race. As Sander himself wrote last week in the Washington Post,“The ‘mismatch hypothesis’ contends that any person (certainly not just minorities) can be adversely affected if she attends a school where her level of academic preparation is substantially lower than that of her typical classmate.”

While Scalia’s defenders contend that he was speaking in shorthand and referencing amicus briefs and scholarly research that convey the foregoing more clearly than he did, I do not think it overzealous “political correctness” to expect more carefully drawn words on this subject in a high-profile hearing, given Scalia’s prominence and the ugly, wrongheaded belief in black inferiority that persists in bigoted enclaves. A man of his position and intellect is capable of better. From a purely consequentialist perspective, he should’ve anticipated that his shorthand would add more heat than light to the debate. As Taylor told the New York Times, “Mr. Scalia’s lack of eloquence had made what he said sound worse than it was.”

If Scalia sometimes shows more talent for provocation than rigor, the press ought to understand how amplifying and denouncing his least careful words misleads readers, who are owed a careful exposition of the actual arguments to which he alluded. At Vox, Libby Nelson at least explained to readers, “Scalia wasn't making up his objection from the bench. He was drawing from a frequent conservative argument against affirmative action: Students with lesser academic qualifications don't benefit from being admitted to a more competitive college.”

But even she gives the impression that mismatch theory is persuasive only to conservative opponents of affirmative action and that it is safely dismissed. As she put it, “research has found this isn't true. If anything, it's the opposite—students benefit from going to the best college that will admit them, even if their academic credentials are a stretch, because more selective colleges tend to have higher graduation rates.”

Sander, who isn’t mentioned in her piece despite his prominence in the debate and an amicus brief in the case before the court, is hardly a conservative ideologue. He graduated from Harvard and chose, as his first job, volunteering as a community organizer on the South Side of Chicago; he wrote his graduate dissertation on fair-housing laws and residential segregation; he worked on the effort to elect Democrat Harold Washington, Chicago’s first black mayor, and served on his transition team; he has a track record of empirical research aimed at figuring out how institutions of higher education can best help struggling students to thrive; and he’s a tenured law professor at UCLA. None of those factors means that his theory is necessarily correct; but they certainly cut against innuendo that “mismatch” is rooted in conservative hackery or racism.

Sander’s initial paper on mismatch theory focused on law schools.

Proponents of mismatch theory––and some agnostics––also cite a study by Duke University’s Peter Arcidiacono, who found that affirmative action might cause its beneficiaries to drop out of the most difficult majors at disproportionately high rates. It is conceivable that such mismatches, if they exist, are particularly likely at law schools and in STEM fields, and less likely to be observed in other disciplines—or even that they vary widely by institution.

Other formidable scholars––Matthew Chingo prominent among them––have offered strong critiques of “mismatch theory.” I have no position to offer on the many questions on which these academics disagree. But perusing the critics, Sander’s most up-to-date defense of his scholarship and various attempts to characterize the overarching debate, I’m baffled that any journalists are treating it as settled, even as tenured social-scientists at top-tier universities declare that it deserves to be taken seriously. No one, it seems, can yet provide a precise answer to the question, “at what point do disparities in GPA, SAT score, or high-school quality start to matter,” even though everyone surely agrees that they matter at some inflection point.

It would be useful to know, whatever it is.

One reason for the stridency of the reaction to Scalia's remarks is the long history of racial exclusion and discrimination in university admissions; some critics fear that “mismatch theory” portends a return to segregated campuses, making the question of mismatches based on race more fraught than it is with respect to preferences based on athletics. Fortunately, almost no one today favors race-based exclusions, and public universities in states like California, where racial preferences are outlawed, have proven able to maintain a large degree of racial diversity. The biggest risk of bungling the mismatch debate is likely leaving students at schools where they are less likely to thrive.

As an observer who is open to the possibility of “mismatch” proponents or skeptics having the better of the argument, the only thing that seems clear to me is that more study of this question is worthwhile. Why does anyone disagree? Nelson writes, “Mismatch theory is always brought up in the context of affirmative action. But universities admit less academically qualified students for all kinds of reasons—because they're the children of alumni or donors, due to athletic or musical talent, and so on. There isn't nearly as much concern about how those students fare, and some research has found they're more likely to drop out than other students, including those admitted through race-based affirmative action.” Wouldn’t it therefore be salutary to have answers to the whole array of questions?

In Mismatch, Sander and Taylor advocate for one step as a possible middle ground. “Schools should provide any information they have available or can reasonably obtain on learning outcomes for past students similar to the admitted student,” they write. “For example, a student admitted to a college with a given SAT score and high-school GPA should receive its best estimate of the past graduation rates of comparable students, their college GPAs, and their rates of attrition from intended majors.”

In one of the most even-handed articles on the controversy, the National Review’s Reihan Salam, an agnostic on racial  preferences, offers a compelling endorsement of that narrow position:

The goal of transparency wouldn’t be to discourage students from attending selective schools. If Chingos is right, the news would in most cases be more encouraging than discouraging. Yet students with below average levels of academic preparation would have a clearer sense of the obstacles they face, and that they’d be wise to take advantage of enrichment resources on campus to keep up with their better-prepared classmates.

It’s not just beneficiaries of racial preferences who’d profit from access information of this kind. So would athletes and benefits of other preferences, like legacy preferences and regional preferences. Indeed, I suspect all students would benefit from having some sense of where they stand in the pecking order. One often hears about students who resent the suggestion that they’ve benefited from racial (or other) preferences. Transparency could do a great deal to address these concerns. If I know that I’ve benefited from a preference, and I’ve made an informed decision about what that will likely mean for my academic prospects, I’d presumably feel far more secure. If I can’t stand the idea of benefiting from a preference, however, I might instead attend a school where I’d start at the top of heap. The ability to make an informed decision in accordance with your values is no small thing.

I suspect information of that sort would’ve better prepared me for the calculus course I struggled through as a Pomona College sophomore admitted with an SAT score that put me well above average in verbal skills and decidedly below it in math. It seems as though it would’ve benefitted Afi-Odelia Scruggs, whose powerful Washington Post op-ed about struggling at an elite college––and benefitting from going there anyway––serves as one powerful anecdotal retort to Scalia’s speculation. And the case for transparency dovetails with the demands of some student activists, who want more transparent data about how successfully institutions eager to recruit them to campus are serving them once they matriculate. Perhaps if lots of private institutions parted with such data, best practices would emerge.

Meanwhile, neither the right nor the left should stigmatize those who disagree with them on this subject. Careful proponents of mismatch theory and its most careful critics are both doing a service by advancing knowledge in an area where accurate information is highly likely to benefit that subset of students of all races who benefit from various admissions preferences, regardless of what the facts turn out to say. Journalists ought to be invested in the rigorous pursuit of that knowledge, and do harm insofar as they overstate what is known, or stigmatize earnest scholars on either side of a debate that won’t be settled without many more studies at diverse institutions.

As John McWhorter put it, this is “a complex matter upon which reasonable minds will differ. With the well-being of young people of color at stake, we can't afford to pretend otherwise.” Stigma and insults help no one. But more knowledge will.