Richard Sander has documented the mismatch effect that occurs when blacks are admitted to selective schools through race preferences. He found that law students admitted under preferences tended to receive lower grades and pass the bar exam at lower rates. Sander posits that without preferences, blacks would be better matched to their schools.
Four Duke University researchers conducted a study in which they argue more information is needed to determine if the mismatch effect exists. Specifically, they contend that mismatch could occur only if selective universities possess private post-enrollment student information, which they call a “necessary condition” to make the determination. (Source)
Rather than looking at objective factors to determine whether the mismatch exists, such as GPAs and scores, the researchers suggest “two potential avenues that may lead to a more conclusive test of mismatch,” which is what they mean by “private post-enrollment information.” The subjective approach includes asking admitted black students what they expected their GPAs to be after their first year and whether they still would have attended if the GPAs turned out to be X.
Confused yet? I think that’s precisely the idea.
Leave it to a group of academics to make something simple seem complicated. As the Center for Equal Opportunity‘s (CEO) Roger Clegg notes, the researchers have in front of them evidence of the mismatch theory already. His comment on NRO’s Phi Beta Cons blog (emphasis added):
All very interesting, but doesn’t this article miss the forest for the trees? The most important point, found in the last line of the table but not commented on, is this: The first-year GPA for white and Asian students was 3.33 and 3.40, respectively; for black students it was 2.90; for Latinos was 3.13. So the ethnic groups that, in all likelihood, got preferences (look at the rest of the table) did substantially worse than the ethnic groups that did not (e.g., a half-point gap between blacks and Asians). Sounds like confirmation of “the controversial ‘mismatch’ theory” to me. (This is not news, of course: The Shape of the River acknowledged that African Americans at selective schools on average had a class rank in the 23rd percentile, versus the 53rd percentile for whites; the Center for Equal Opportunity’s studies over the years have found similar gaps.)
Such lopsided GPAs (not to mention test scores) is evidence of mismatch, isn’t it? Why is it necessary to send letters to students asking them to speculate on first year grades? The mismatch isn’t subjective; it’s an effect that either exists or doesn’t exist. If students admitted under preferences have substantially lower grades than students not admitted under the policy, there is a mismatch.
Download “Does Affirmative Action Lead to Mismatch? A New Test and Evidence” here. (PDF – 33 pages)
Comments on this entry are closed.