Failed Effort to Prove Diversity

“Diversity is good and we have the proof!” say advocates of racial preferences in college admissions. But statistics can be made to prove just about anything if you are willing to confuse vague feelings with facts.

For years, there has been an academic industry devoted to justifying college “affirmative action” — the policy of giving preferences to some students over others based on group identification. Allegedly, that is necessary to achieve “diversity” on campus – a student body reflecting various racial and ethnic groups, particularly African-Americans and Hispanics. (People are different in far more ways than race and ethnicity, but to diversity proponents, only a few of those ways matter.)

Diversity advocates keep hunting for evidence that there are educational benefits from their policy of admissions preferences because the Supreme Court has ruled that preferences are unconstitutional unless they serve some compelling interest.

A paper published last March purports to find proof that diversity yields educational benefits. Authored by Steve Chatman, a researcher at the University of California, the paper “Does Diversity Matter in the Education Process?” has been extolled by Wake Forest professor Joseph Soares, who says (scroll down) that it is a “solid empirical study” demonstrating that diversity teaches students “to broaden their perspective and sensibilities.”

Does the paper really do that? No, but knowing what’s wrong with this supposed proof of the benefits of diversity is important if we are ever going to accurately assess the pros and cons of this contentious policy.

The data that Chatman analyzes are questionnaires completed by a large number of students at the campuses in the University of California system. Students were asked:

    How often have you gained a deeper understanding of other perspectives through conversations with fellow students because they differed from you in the following ways?

    Their religious beliefs were very different than yours.
    Their political opinions were very different from yours.
    They were of a different nationality than your own.
    They were of a different race or ethnicity than your own.
    Their sexual orientation was different.

Chatman says that the responses from students give “useful, if soft evidence of diversity benefits.” Unfortunately, this evidence can’t even be called soft. It’s nonexistent. Here’s why.

  • First, self-reports are notoriously unreliable data. Many people are prone to saying what they think they ought to say on surveys. That is especially true here since, as Chatman admits, most of the students in the sample identify themselves as Democrats or learning toward the Democrats. That matters because the Democratic Party is strongly identified as favoring “affirmative action.” Therefore, we should suspect that at least some students would give answers supporting diversity just because they have heard from politicians they like that affirmative action is good. It’s like asking a mostly Republican group of students if they think the “surge” in Iraq was a success.

    Also, for students who aren’t Democrats, answering that they had few or no “diversity experiences” leading to “deeper understanding” might give the impression that they aren’t sensitive individuals. That is something many people want to avoid, even on anonymous surveys.

  • Second, note that the question is phrased in a way that leads to positive views about diversity. Diversity experiences are couched in language implying that they necessarily lead to “deeper understanding.” The only question is how often that happens. Lawyers would call that “leading the witness.” Logically, it’s possible that students might have diversity experiences that don’t matter at all, or even lead to antagonism, but they can’t report that.
  • Third, exactly what counts as “deeper understanding”? Interpreting that vague phrase was up to the students. Even a trivial increase in knowledge about a “different” person might be thought to qualify. If, for example, a Christian student learned through a “diversity experience” something about Islamic dietary customs, would that be “deeper understanding”? Perhaps, but would that have any real importance? Or if a black student learned that a white student liked opera, would that be an advance for cultural understanding, or just knowledge about a personal preference?

All that we can say for sure after looking at this data set is that a large number of UC students feel good about campus diversity. It tells us nothing at all about any changes in their behavior as a result of their “diversity experiences.” Diversity proponents make the assumption that it promotes “deeper understanding,” thus making society more harmonious, but these questionnaire responses offer no evidence of that. It’s quite possible that “diversity experiences” make no difference in real life.

Assume for example that a white student answered that he had many of these experiences regarding race. Assume also that he was perfectly comfortable with and accepting of people from other races – as is usually the case with young Americans. Granting that he actually has gained “deeper understanding” of people from other cultures, so what? He was already well adjusted to the fact that people are different.

This study provides no evidence that diversity experiences in college have lasting, beneficial effects, but Chatman leaps to the conclusion that they do. He writes, “A stimulating environment of interchange among students that will help them succeed after college requires that the student body exhibit diversity in areas important to society. If students are to function effectively in a world with immigrant, political, religious, social class, and racial/ethnic differences, then the student body should include students with a variety of these characteristics.”

That’s a glaring non sequitur. The student responses Chatman analyzes do not tell us anything about the ability of graduates to “function effectively” in society. There isn’t an iota of evidence presented to show that students who have attended colleges with greater student body diversity are better at dealing with people than are students who graduated from colleges with little or no racial diversity. (Virtually every college has plenty of political, religious, and socio-economic diversity.) Chatman’s conclusion expresses the wish of diversity advocates, but his paper does nothing whatever to prove it.

He tries to shore up the paper’s weakness by pointing to statements by business executives that they favor admissions preferences for “underrepresented minorities.” In an amicus brief in Grutter v. Bollinger, General Motors asserted that it needs a work force “comprising people who have learned to work productively and creatively with individuals from a multitude of races and ethnic, religious, and cultural backgrounds….” Chatman attaches great significance to that statement, never considering the possibility that for General Motors (and other businesses), professing support for affirmative action is an inexpensive way of avoiding harassment from people like Jesse Jackson.

Notice that General Motors doesn’t claim that workers who attended colleges with little or no racial/ethnic diversity have proven themselves unable to work productively and creatively in its workforce. Has G.M. hired white students from, say, North Dakota State and black students from, say, North Carolina A&T and found that they aren’t good at cooperating with people who are “different”? If so, that would at least be weak evidence in favor of using racial (and other) quotas in college admissions.

To my knowledge, however, no evidence linking workplace effectiveness to the composition of the student body of an individual’s alma mater has ever been adduced. We shouldn’t hold our breath waiting for it, for the simple reason that workplace cooperation does not require deep cultural understanding among workers. Toleration suffices and most people have enough of that to get by. Highly diverse American workplaces functioned smoothly long before colleges started worrying about “diversity.”

Those who demand college “diversity” overlook the truth that every college campus is diverse. Human beings, even from the “same” group, are different in a hundreds of ways. Two white students might differ more than a white and a black student. Furthermore, Americans encounter ample “diversity” in society long before getting to college. The notion that Americans would have no “diversity experiences” if colleges didn’t employ racial preferences to ensure quotas of certain groups is absurd.

The alleged benefits of racial preferences to achieve “diversity” in higher education have yet to be shown, but there is strong reason to believe that they have done considerable damage. See, for example, Professor John Ellis’s essay “ “TARGET=”_blank”>How Preferences Have Corrupted Higher Education.” If the intellectual case for “diversity” is no better than jumping to illogical conclusions based on feeble survey evidence, it is time for proponents of racial preferences to admit that this dispute is about politics rather than education.