February 14, 2007: Features

Getting in ... and what comes later
A former president speaks on diversity in higher education

By William G. Bowen *58

At Princeton’s “Kaleidoscope” conference on race and community in November, former president William G. Bowen *58 spoke to alumni about diversity in admissions. His talk, which spanned about 40 years of Princeton history and detailed his current research related to race, income, and college admissions, was especially well received by alumni. PAW presents an excerpt here.

Bowen began by recalling the years when Princeton began to recruit African-American students, “at first, rather timidly, and then more and more aggressively.” The Class of 1971, he recalled, had only 14 or 15 African-Americans, but he described their impact as profound. “In retrospect,” Bowen said, “we clearly underestimated how hard it would be to overcome all the factors that made minority students feel unwelcome and less than a full part of Princeton.”

Later, Princeton also recruited students from other minority backgrounds more actively. Bowen recalled how he convinced Gordon Wu ’58 to allow the University to inscribe his name in Chinese calligraphy at the entrance to Wu Hall. “The argument I used in convincing Gordon to let us place the calligraphy in its proper place was that this would be a way of symbolizing that those who care about Princeton, and provide for it, are by no means only the donors with ‘traditional’ names like Pyne and Dillon,” Bowen said.

He spoke of the beginnings of coeducation, noting that by 1977, women were so much a part of the University that a woman in the freshman class did not realize Princeton had ever been all-male. And he described religion as a “more important aspect [of inclusiveness] than some realize,” noting that through the 1980s, Princeton “still seemed to say, in many direct and some subtle ways, that Jewish students, like the minority students, were ‘outsiders,’ at Princeton only on Princeton’s terms.” This changed, he said, with the decision to move important ceremonial events like Baccalaureate outside Christian worship services and with the opening of the Center for Jewish Life.

We pick up Bowen’s address as he discusses his current research, about race and socioeconomic status in college admissions:

In the late 1990s, [former Harvard president and now interim president] Derek Bok and I embarked on a major empirical study of the actual effects of race-sensitive admissions policies, as they could be discerned through a detailed analysis of a database built at the Mellon Foundation between 1995 and 1997. The database contained the in-college records of basically all matriculants in the student cohorts that entered 28 academically selective colleges and universities, including Princeton, in 1951, 1976, and 1989.

In light of the increasingly contentious public debate over the role played by race in admissions, legislative actions in California and the state of Washington prohibiting any consideration of race, the sweeping Hopwood decision in Texas that threatened to overturn Bakke, and the strong likelihood that there would be a further Supreme Court review of affirmative action, the time seemed right for a comprehensive empirical review of how these policies had actually worked. We sought to “find the facts,” as best we could. I remember a Princeton trustee and a lifelong friend, Nick Katzenbach ’43, telling me prophetically at the time we began this research: “Remember, you are writing this book for one person: Justice Sandra Day O’Connor.”

Our findings were reported in The Shape of the River: Long-term Consequences of Considering Race in College and University Admissions (1998), and I will now summarize the principal lessons learned through this research, the debate engendered by it, and follow-up and companion studies. Five propositions stand out:

First, the educational benefits of diversity have been strongly affirmed. Patricia Gurin and her colleagues at the University of Michigan assembled and assessed masses of evidence (including the findings in The Shape of the River) on the effects of diversity on both “learning outcomes” and “democracy outcomes.” Gurin concluded: “We have documented a consistent picture from both our research and the research of other scholars that shows a wide range of educational benefits when students interact and learn from each other across race and ethnicity.”

Second, race-sensitive admissions policies have increased substantially the number of well-prepared minority students who have gone on to assume positions of leadership in the professions, business, academia, the military, government, and every other sector of American life — thereby reducing somewhat the continuing disparity in access to power and opportunity that is related to race in America. The evidence presented in The Shape of the River shows that minority students admitted to academically selective colleges and universities as long ago as the mid-1970s not only have graduated at very high rates, but have completed rigorous graduate programs, done well in the marketplace, and, most notably, contributed in the civic arena out of all proportion to their numbers.

Third, the costs of race-sensitive admissions policies have been modest and are well justified by the benefits; most of the alleged negative effects of race-sensitive admissions are seen, on examination, to be minor or nonexistent. Specifically, there is no evidence that, overall, race-sensitive admissions policies “harm the beneficiaries” by putting them in settings in which they are overmatched intellectually or “stigmatized” to the point that they would have been better off attending a less selective institution. This assertion withers in the light of the evidence. The most compelling and relentlessly consistent data in The Shape of the River show that, far from being stigmatized and harmed, minority students admitted to selective colleges under race-

sensitive admissions performed very well. Indeed, the more selective the college they entered (holding their own SAT scores constant), the more likely they were to graduate and earn advanced degrees, the happier they said they were with their college experience, and the more successful they have been in their careers. This is precisely the opposite result from the one that proponents of the “harm-the-beneficiary” line of thinking would have predicted.

Evidence also disposes of the argument that substituting race-neutral admissions policies for race-sensitive policies would have removed from campus a marginal group of mediocre students, leaving only a distinctly superior “top tier” of well-qualified minority students. Examination of the later accomplishments of those students who would have been “retrospectively rejected” under race-neutral policies shows that they did just as well as a hypothetical reference group that might have been accepted if admissions officers had given black students and white students with the same GPAs and test scores exactly the same consideration. There are no significant differences in graduation rates, advanced-degree attainment, earnings, civic contributions, or satisfactions with college. These striking results testify, we believe, to the excellent job done by admissions officers in “picking and choosing” among the large number of applicants from underrepresented minority groups as well as other candidates who are well over the admissions threshold — many of whom are rejected.

Contrary to what is often supposed, eliminating race-sensitive admissions policies entirely would have increased the admission rate for white and Asian applicants at these academically selective institutions by less than two percentage points: from roughly 25 percent to 26.5 percent. At all selective colleges and universities with a plethora of well-qualified candidates, the “opportunity cost” of admitting any particular student is that another strong applicant is not chosen. A rejected applicant — and the applicant’s parents — sometimes assume that he or she surely would have gotten in had it not been for “reverse discrimination.” In fact, race-sensitive admissions policies for underrepresented minorities have not reduced appreciably the chances of other well-qualified applicants to gain admission to the most selective colleges and universities — in many situations, recruited athletes receive larger admissions “breaks” and displace more applicants than do minority students.

Fourth, progress has been made in narrowing test-score gaps between underrepresented minority students and other students, but gaps remain — and so does the need for race-sensitive admissions policies. At a group of liberal arts colleges and universities examined in 1976 and 1995, average combined SAT scores for underrepresented minority students rose roughly 130 points at the liberal arts colleges and roughly 150 points at the research universities. Test scores of other students rose too, but by much smaller amounts. In short, test-score gaps narrowed over this period, and the average rank-in-class of minority students on graduation improved even more than one would have predicted on the basis of test scores alone. Still, more progress needs to be made. At the 10 institutions for which we have limited data on the 1999 entering class, there was a 137-point gap between the SAT scores of minority students and non-minority students.

Finally, one unanticipated consequence of the publication of The Shape of the River is that a number of African-American students who attended the selective schools included in the study found the highly positive findings very reassuring — they served to counter hurtful suggestions that these students did not really deserve to be there. After I made a presentation to the American Council on Education, a young black woman came up and introduced herself as a Harvard graduate. She said that she found the book “liberating” and added: “I guess I have been walking around with a kind of cloud over my head, which I didn’t really understand was there, and it’s gone now. ... We did pretty well, didn’t we?” “Yes,” I responded, “you did very well indeed.” She nodded and said, with emphasis, “Thank you. I’m now ready to stand up and fight!” Also, I remember well a call I received in my office from Jerry Blakemore ’76. who had been in my Economics 101 section, and who said simply, in commenting on The Shape of the River, “President Bowen, your book is about me!”

I turn now to socioeconomic status. Following publication of The Shape of the River and two other studies focused on athletics, colleagues and I completed another large empirical study, focused this time on the efforts of 19 academically selective colleges and universities (again, including Princeton) to educate large numbers of students from modest socioeconomic-status backgrounds.

What did we find? In brief:

First, aggressive outreach and recruitment efforts notwithstanding, only about 11 percent of the students at these schools were from the lowest income quartile, only about 6 percent were first-generation college-goers, and only about 3 percent were first-generation college-goers who also came from the bottom income quartile.

Next, cutting through a mass of analysis, one main take-away is that, for those applicants who took the SAT, did reasonably well on it, and applied to one of the selective institutions, family income and parental education in and of themselves had surprisingly little effect on admissions probabilities, yields, choices of majors, and subsequent academic performance and graduation rates. Given the importance to American life of reducing disparities related to family circumstances, of increasing mobility, and of increasing the country’s stock of human capital in an ever more competitive world, I don’t think this is an outcome that is good enough. I am proud to say that President Tilghman and Princeton have taken the lead in improving financial-aid packages and in enrolling more students from modest backgrounds. This is unfinished business.

We also spent considerable time studying the interactions between race and socioeconomic status [SES]. The key question is whether focusing admissions and enrollment efforts solely on socioeconomic status would simultaneously, as a kind of by-product, address the need for racial diversity. Since it is true that a far larger fraction of blacks than of whites come from the bottom income quartile, it is easy to see why some people might be intrigued by this way of avoiding the issue of race. Americans have an insatiable appetite for the painless solution, and considering race continues to be painful for many people.

We carried out some complex simulations that allowed us to estimate the effects on minority enrollment at these schools of substituting for race-sensitive admissions one fairly aggressive way of doing more for candidates from low-income backgrounds (giving them the same admissions advantage as is given to legacies today). The numbers are not encouraging. Replacing race-sensitive admissions with this kind of “legacy thumb on the scale” for low-income students would yield student populations with a few more minority students than would be there if neither race-sensitive admissions nor the SES “thumb on the scale” were used. But the comparison with the outcome of a race-sensitive policy is sobering: Substituting our SES “thumb” for race-sensitive admissions would cut the number of underrepresented minority students in the typical undergraduate class at these institutions roughly in half: from just over 13 percent to 7 percent. The effects at the graduate and professional levels would be even more dramatic.

This last set of findings has direct relevance for an extremely timely debate: the future of affirmative action. You probably saw that [in November] voters in Michigan approved a ban on essentially all kinds of affirmative action by public institutions, including universities, in that state. Colleagues at the University of Michigan are both disappointed and dismayed by this result, which could have very serious consequences for the ability of that great university to continue to enroll a racially diverse student body, especially in its graduate and professional schools.

My own current research is focused on educational attainment at the leading public universities, and especially the contribution these universities are making in educating more students from both modest economic circumstances and racial minorities. The Michigan results give me a new incentive to press on with that research as vigorously as possible. The outcome of the Michigan referendum is sobering for those of us who believe, with [legal philosopher] Ronald Dworkin, that “the worst of the stereotypes, suspicions, fears, and hatreds that still poison America are color-coded.”

In speaking at a conference on educational opportunity at the University of North Carolina, I referred to the last speech that President Lyndon Johnson gave, just about a month before he died, at a symposium on civil rights in Austin, Texas. He was desperately ill; he had been told by his doctor not to attend the symposium and certainly not to subject himself to the strain of giving a major speech. But LBJ was determined. Let me conclude by quoting an excerpt from that talk: “To be black, I believe, to one who is black or brown, is to be proud, is to be worthy, is to be honorable. But to be black in a white society is not to stand on level and equal ground. While the races may stand side by side, whites stand on history’s mountain, and blacks stand in history’s hollow. We must overcome unequal history before we overcome unequal opportunity.”

This conference is for me a powerful reaffirmation of Princeton’s recognition of the importance of not forgetting where we have come from as we focus on what is clearly an unfinished agenda.