THE TRUTH ABOUT PREFERENCES
Jul 21, 1997, Vol. 2, No. 44 • By GAIL HERIOT
A FEW YEARS AGO, no secret was more closely guarded than university affirmative-action statistics. "Preferences? What preferences?" was the official line, and the occasional leak was handled with astonishing ruthlessness. Consider the case of the law student who exposed the credentials gap between "affirmative action" and "regular" admittees at Georgetown University Law Center. Only a public outcry caused Georgetown to back down from its threat to expel him. Instead, university administrators satisfied the calls for blood on campus by reporting his conduct to the bar. All the while, Dean Judith Areen blandly maintained that Georgetown did not discriminate on the basis of race.
Things have changed; now we're awash with information. It seems that every week there are more press conferences, more news releases, more National Public Radio interviews.
The good news is that universities are finally admitting what most people knew all along, even though it was difficult to pry loose specific information: In campus admissions offices, race matters. A lot. Those of us who wielded the crowbar in years past are gratified by this newfound candor.
But candor is actually just a tactical retreat for the defenders of preferences. They now claim that the beneficiaries of racial preferences perform just as well as the better-qualified students. Hence those who call for a greater emphasis on academic qualifications and less emphasis on skin color are anguishing over nothing -- or so the argument runs.
In truth, the data prove just the opposite. But the media have reported the claim uncritically. And with each retelling, the real facts become more and more distorted. Tim Russert put it this way on Meet the Press: "A study done . . . of law school admissions in 1991 . . . concluded that blacks who were admitted under affirmative action performed just as well as whites in law school and on passing the bar exam."
Wrong. In California, when it came to passing the bar, the gap between whites and African Americans was so wide that the bar examiners modified the test format to include a "performance test" of practical skills, hoping that it would close the gap. It didn't. In 1994, well after the addition of the skills test, the bar passage rate for first-time takers in California was 81. 3 percent for whites, 53 percent for African Americans.
In New York, the disparity was greater. The Millman Report, commissioned by the New York Court of Appeals to examine this and other problems, found that whites pass at a rate of 81.6 percent, as opposed to 37.4 percent of African- American test-takers.
This sorry picture is almost solely the result of preferences. When African- American exam-takers compete against white exam-takers who have the same score on the law-school admissions test, they fare about the same, according to Stephen P Klein, a research scientist for the Rand Corporation. But a system of preferences necessarily pits many minority students against students with substantially better academic credentials. The result is predictable.
The problem is not unique to law schools. A study published in the Journal of the American Medical Association reports that white medical students passed the National Board of Medical Examiners Part I exam at a rate of 87.7 percent, African Americans at 48.9 percent. Again, matching academic credentials (in this case, the MCAT score and undergraduate GPA) made the discrepancy all but disappear.
So what is the new report Russeft referred to, and how could it reach such an unexpected conclusion? Polemically titled "The Threat to Diversity in Legal Education," the report was written by Linda Wightman and published in the April issue of New York University Law Review. Contrary to Russert's representation, the study does not even purport to compare white student performance with African-American performance. It compares exclusively within race. AfricanAmericans who would not have been admitted without a preference are compared with African-Americans who did not need one. Similar comparisons are made for other racial and ethnic groups.
The study does not find that, among African Americans, beneficiaries of preferences performed "just as well" as non-beneficiaries. To the contrary, it reports that AfricanAmerican preference beneficiaries failed the bar at a rate almost three times that of non-beneficiaries of the same race (27.1 percent vs. 9.8 percent) -- a highly significant statistical difference. This is hardly performing "just as well."