MISMATCH: HOW AFFIRMATIVE ACTION HURTS STUDENTS IT'S INTENDED TO HELP, AND WHY UNIVERSITIES WON'T ADMIT IT
By Richard H. Sander and Stuart Taylor Jr.
Basic Books, $28.99, 348 pages
Conservatives are always looking for their holy grail of social science: empirical proof that liberal policies do more harm than good. If Charles Murray was right in "Losing Ground" and welfare actually makes the poor worse off, the debate is over -- no one wants to do that.
For years now, Richard H. Sander of UCLA has been floating around a theory of this type for affirmative action. In "Mismatch," co-authored by journalist Stuart Taylor Jr., Mr. Sander brings together all the evidence he can find that racial preferences don't help the people they're supposed to help.
The idea of mismatch is not hard to understand. Some colleges are more selective than others, and colleges that are more selective also tend to be more rigorous. At top colleges, professors assume students are prepared to keep up with a steady stream of challenging new material.
Usually, this assumption is correct, because university admissions departments are good at picking the very best applicants. However, about 25 percent of college students attend schools that use racial preferences. (The other schools typically are not very selective to begin with.) At the most selective institutions, racial preferences tend to be incredibly strong, to the point that there isn't much overlap in test scores between black students and white ones. Hispanics tend to get less preference than blacks, and Asians are discriminated against, sometimes severely.
Affirmative action, in other words, is taking black and sometimes Hispanic students who are perfectly qualified to attend college and placing them in schools where their peers are much more advanced. One would imagine these students would get lower grades than their peers, would be more likely to drop out and would be more likely to leave challenging majors for softer ones that won't be as useful in the job market.
Great theory, but where's the proof? That's the big problem: Academia is notoriously cagey when it comes to racial preferences. If a critic says affirmative action is unfair, academia replies that it's not so bad -- it's just a "tiebreaker," or maybe a little "boost." If a critic suggests eliminating preferences, academia replies that this would destroy minority enrollment numbers, suggesting that preferences are quite strong indeed.
If a critic asks for the data, academia tells him to shove it. It's often difficult to tell how strong racial preferences are at any given school, let alone how students admitted under preferential policies fare during their four years on campus. As a result, "Mismatch" can feel cobbled together at times. Most readers will long for a comprehensive data set to end the debate.
Nonetheless, what Mr. Sander and Mr. Taylor have accomplished here is incredibly impressive. The authors have done an excellent job of pulling together the available research, and Mr. Sander in particular has been dogged in his pursuit of fresh numbers. He's even used lawsuits and Freedom of Information Act requests to extract data from admissions departments.
Mr. Sander summarizes the excellent research he's done on law schools, for example. Black students admitted under racial preferences tend to drop out at much higher rates than black students who attend schools that match their qualifications, and black law school graduates are far more likely than whites to fail their bar exam. By Mr. Sander's estimation, more blacks would succeed in becoming lawyers if they attended schools that would accept them without preferences.
A similar effect also seems clear in STEM fields (science, technology, engineering and math). While minorities are just as interested in these fields as whites, they fail disproportionately when they attend selective schools. The data show that in these fields, a student's preparation relative to that of his peers is especially important -- a student who is capable of flourishing at a middle-tier school may very well fail if he's admitted to Massachusetts Institute of Technology.
The authors also take a look at the effects of Proposition 209, the California ballot initiative that briefly eliminated racial preferences in the University of California system. (Prop 209 is still in effect, but schools have found ways to get around it.) When Prop 209 first went into effect, preference supporters pointed out declining minority enrollment, particularly at the University of California-Los Angeles and the University of California-Berkeley.
A deeper look at the data tells a different story. Minority enrollment rose at three of the system's campuses and hardly changed at two more. Minority graduation rates swelled -- more than canceling out the effect of lower admissions, meaning the system was graduating more minorities than it was previously. Minority applicants who were accepted were more likely to attend than they had been. Also, applications from blacks and Hispanics actually rose, suggesting they were not dissuaded by the lack of preferences.
Mr. Sander and Mr. Taylor, of course, have their share of critics, and "Mismatch" will not be the last word on this subject. But they have put the nation's universities in a put-up-or-shut-up situation: They can either admit that preferences do harm, or they can release the data that prove otherwise.
Robert VerBruggen is a deputy managing editor of National Review.