RESEARCH IMPLIES DIVERSITY MIGHT NOT BE SO DESIRABLE

Saturday, April 5, 2003

The original rationale for racial-preference policies like those at the University of Michigan was to compensate for the results of past societal discrimination. That's what U.S. Supreme Court Justice Lewis Powell said in the 1978 Bakke case when he provided a crucial fifth vote for the position that universities could take race into account in making their admissions decisions.

Over time, however, that argument has largely been abandoned, in part because it was so hard to explain why preferences were not available to students of Chinese or Japanese ancestry, some of whose families could have been subject to legal discrimination in America, while they were available to certain students, not all, whose families came to America after the passage of the Civil Rights Act that eliminated legal discrimination.

The favored excuse now is one grounded in the virtues of diversity, stressing the benefits -- to white students, primarily -- of going to college with people of other races.

To most people it seems obvious that it's a good thing for college students to get to know people unlike themselves in a whole variety of ways, race being one. The more diversity, the better. The question rather is on the morality and the constitutionality of the means chosen to accomplish a specific level of racial diversity. The way Michigan did it was by admitting minority students with significantly weaker academic credentials, on average, than white or Asian students. Michigan's policies are the subject of two cases now before the Supreme Court, which heard oral arguments Tuesday.

But is it really true that more diversity is better? Hardly anybody has studied that question, but research recently published by Stanley Rothman, Seymour Martin Lipset and Neil Nevitte suggests the answer might be ``no.'' Their article appeared in the spring issue of the International Journal of Public Opinion Research.

The authors surveyed students, faculty and administrators from 140 predominantly white U.S. colleges and universities. Their structured random sample matched the college-attending population in race and gender.

A lot of the research in this area relies on asking people what they think about diversity, and either because they do believe it's a good thing or because they know they are supposed to believe it's a good thing, they say they approve of it.

For this survey, the authors asked general questions about the campus climate -- for all groups, how well the school educates its students and how hard students work; for students only, how satisfied they were with their campus experience; and for faculty and administrators only, how well prepared academically students were when they arrived on campus.

They also asked -- these questions came later in the survey, Rothman said -- about how minorities are treated on campus, whether the person responding had ever been treated unfairly, and the extent to which racial discrimination is a problem on campus. Then they correlated the results with an independent measure of diversity, the proportion of African-American students on campus (later papers will look at other minority groups).

The diversity hypothesis being tested was that ``increasing black student enrollment in predominantly white student bodies will produce a better educational environment, greater attention to and satisfaction with the quality of education, and better relations between white students and students of color.''

Well, guess what? Four of the six questions asked of students showed statistically significant results and all four went in the opposite direction from that optimistic prediction. As black enrollment rose, students' satisfaction dropped along with their opinion of the quality of education and their fellow-students' work efforts. And they were more likely to say they, themselves, had experienced discrimination.

Faculty and administrators, likewise, had five statistically significant negative correlations of a possible six on the questions relating to academic climate, and three positive correlations (one weak) of six relating to treatment of minorities and racial discrimination.

This is very bad news, and not only for Michigan's case. But you shouldn't blame the authors for bringing it. Research shows what it shows, even when it is unexpected and unwelcome.

The researchers carried out a further analysis to examine to what extent other variables -- individuals' gender, economic status and major subject, as well as characteristics of the institutions -- contributed to the results. Enrollment diversity, they write, ``contributed significantly to explaining the variance in students' evaluations of college life, after controlling for all other demographic, academic and institutional factors.''

Perhaps these discouraging results are caused not by the presence of black students as such, but rather by the vigorous efforts of universities such as Michigan to enroll black students even if they are less well prepared on average. The law school at Michigan, after all, conceded that nearly three-fourths of the black students it enrolled wouldn't have been admitted without preferences.

Overall, though, this research suggests that the push for diversity might do more to create stereotypes than to break them down. Attention must be paid.