Book Search:  


Google full text of our books:


The Shape of the River:
Long-Term Consequences of Considering Race in College and University Admissions
William G. Bowen & Derek Bok
With a new introduction by the authors and a foreword by Glenn C. Loury

Book Description | Reviews | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 1998, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format


Stretching from St. Paul to New Orleans, Mark Twain's Mississippi winds for twelve hundred miles through fog, rapids, slow eddies, sandbars, bends, and hidden bluffs. Drawing upon his own experiences on the Mississippi, Twain created an image of the river as both physically central to the United States and symbolically central to the progress of the country. The image of the river is also central to the story of our book, which is concerned with the flow of talent--particularly of talented black men and women--through the country's system of higher education and on into the marketplace and the larger society.

    The image most commonly invoked in discussions of this process is the "pipeline." We often hear of the importance of keeping young people moving through the "pipeline" from elementary school to high school to college, on through graduate and professional schools, and into jobs, family responsibilities, and civic life. But this image is misleading, with its connotation of a smooth, well defined, and well understood passage. It is more helpful to think of the nurturing of talent as a process akin to moving down a winding river, with rock-strewn rapids and slow channels, muddy at times and clear at others. Particularly when race is involved, there is nothing simple, smooth, or highly predictable about the education of young people.

    While riverboat pilots on the Mississippi navigated "point to point"--only as far as they could see into the next bend--they had to know every depth, every deceptive shoal, and every hidden snag of the river. Moreover, since the boats ran throughout the night, in high water and low, and both up the river and down it, these pilots had to know the river's features in every imaginable condition, and from either direction. Even though they could only steer through what they saw in front of them, they had to understand how the bend that they were navigating at any moment fit into the shape of a twelve-hundred-mile river.

    The college admissions process and the educational experience that follows it are similarly complex. Most recently, debate about the use of race as a criterion has centered on the question of who "merits" or "deserves" a place in the freshman class. At this one bend in the river, prior grades and numerical test scores offer a tempting means of defining qualifications, since they are easily compiled and compared. But what do they really tell us, and what are we trying to predict? Much more, surely, than first-year grades or even graduation from one college or another. It is the contributions that individuals make throughout their lives and the broader impact of higher education on the society that are finally most relevant.

    In this book, we seek to be helpful to both the "pilots" of this educational process--the parents of prospective students, the high school counselors, college admissions officers, faculty members, and administrators, trustees, and regents responsible for setting policies--and those future students who will some day have to navigate the river. We also hope this study will be useful to employers, legislators, and the public as a whole, since everyone has an interest in the development of talent and access to opportunity in our society. We need to know as much as we can about what has happened around bends and curves--in college, in graduate school, and then twenty years downstream--from the frozen moment in time when seventeen-year-olds from various races and backgrounds sat down with Number 2 pencils to take the SAT. This book is an attempt to chart what race-sensitive admissions policies have meant over a long stretch of the river--both to the individuals who were admitted and to the society that has invested in their education and that counts so heavily on their future leadership.

    These questions are enormously important because this country is not yet where any of us would want it to be in terms of race relations. On this central point, liberals and conservatives often agree. Echoing W.E.B. Du Bois, John Hope Franklin has argued eloquently that "the problem of the twenty-first century will be the problem of the color line. ... By any standard of measurement or evaluation the problem has not been solved in the twentieth century, and thus becomes a part of the legacy and burden of the next century." The problem of "the color line" is so central to American life for reasons that are rooted in the disjunction between the values embedded in the Constitution and the realities of three centuries of collective experience. These reasons reflect a sense on the part of many that, despite all the progress made in the past fifty years, we have not yet succeeded in transcending a racial divide that too often discourages the development of ordinary relationships among individuals based on trust and mutual respect. They include as well persistent gross inequities in wealth, privilege, and position that are hard to explain away simply on the basis of differences in individual effort and initiative, significant as such differences are. Finally, there is a collective concern that we are failing to develop to its fullest the human potential of the country and a growing realization that our society, with its ever more diverse population, cannot ultimately succeed as a democracy if we fail to close the gaps in opportunity that continue to be associated with race.

    The subject of race in America is as sensitive and contentious as it is important. Highly charged words, such as "fairness," "merit," "achievement," "preference," and even "race" itself, often take on very different connotations depending on the speaker and the context. (Note, for example, the radical differences in polling results when the wording of questions about race is changed in relatively minor ways.) Language itself has been a casualty of heated debate; for this reason one aim of this study is to "unpack" the meaning of terms such as "merit," clarify their various possible meanings, and set forth the consequences of embracing one conception of what they signify rather than another.

    Our country respects individual achievement, but it also recognizes that what people have achieved often depends on the families they have grown up in, the neighborhoods in which they have lived, and the schools they have attended, as well as on their own ability and hard work. People rightly seek a society in which racial prejudice no longer limits opportunities. But any close observer of American society cannot help but see the many ways in which, covertly and overtly, consciously and unconsciously, actively and as a consequence of inertia, racial differences that have been long in the making continue to thwart aspirations for an open and just society. Words reflect this reality. When an interviewer interested in nomenclature asked the distinguished social psychologist, Kenneth Clark, "What is the best thing for blacks to call themselves," Clark replied: "White."


Many Americans are uncomfortable about the use of race as a factor in admitting students to selective colleges and professional schools. Critics have attacked the policy on several grounds. They maintain that it is wrong for universities to exclude white applicants with high grades and impressive test scores while accepting minority applicants with lower grades and scores. They point out that admissions officers sometimes accept minority applicants who are not disadvantaged but come from wealthier, more privileged homes and better schools than some applicants who are rejected. They claim that all such policies accentuate racial differences, intensify prejudice, and interfere with progress toward a color-blind society. They assert that admitting minority applicants with lower grades and scores may stigmatize and demoralize the very students that the policy attempts to help, by forcing them to compete with classmates of greater academic ability.

    Defenders of race-sensitive admissions respond with arguments of their own. They insist that such policies are justified to atone for a legacy of oppression and to make up for continuing discrimination in the society. They point out that admissions officers have long deviated from standardized test scores and prior grades to favor athletes, legacies, and other applicants with special characteristics that are deemed desirable. They argue that admitting a diverse class gives students of all races a better preparation for living and working in an increasingly diverse society.

    Until now, the debate has proceeded without much empirical evidence as to the effects of such policies and their consequences for the students involved. The chapters that follow seek to remedy this deficiency by drawing on an extensive study of students from a number of academically selective colleges and universities--places where the debate over race-sensitive institutions has been played out in "real time." We are concerned primarily with the performance, in college and after college, of black and white students admitted to these schools.

    In setting forth the "facts," as best we can discern them, we recognize that all data of this kind are subject to many interpretations. Moreover, even considering such questions can antagonize people on both sides of the argument who believe that the "right principles" are so compelling that no amount of evidence can change their minds. Plainly, data take us only so far in considering this subject. Individuals who agree on "the facts" may still end up disagreeing about what should be done because of overriding differences in values. As a result, we have no expectation that the analyses presented in this study will resolve complex issues to everyone's satisfaction. But we do hope that our research can inform the debate by framing questions carefully and presenting what we have learned about outcomes.

    Of course, it is widely understood that in framing questions and testing hypotheses, investigators are always influenced by their own values and preconceptions. We know that we have been. It would be disingenuous not to acknowledge that both of us came to this study of race-sensitive admissions with a history of having worked hard, over more than three decades, to enroll and educate more diverse student bodies at two of the country's best-known universities. This does not mean that we have favored quotas (we have not) or that we are unaware of how easy it can be for good intentions to lead people astray. Nor have we ever believed that all colleges or universities--including those with which we have been most closely involved--have always made the right choices or implemented every policy perfectly. Still, the fact remains that we are both strongly identified with what we regard as responsible efforts to improve educational opportunities for well-qualified minority students.

    At the same time, in contemplating this study, we recognized that race-sensitive admissions policies rested on a set of assumptions that had not heretofore been tested empirically. Much basic information was lacking about such topics as the academic performance of minority students with higher and lower test scores in the most selective colleges and universities, the nature and extent of interaction among different races on campus, and the subsequent careers of minority students accepted through race-sensitive policies. When we began the study, we were far from certain what the data would reveal. Quite possibly, some important assumptions underlying the efforts to enroll more minority students would turn out to be unfounded. Nevertheless, we felt that after thirty years, it was surely time to discover the facts, insofar as it was possible to do so. It was important, we thought, to try to understand and come to terms with any disappointing results as well as to learn from positive outcomes. Now that we have completed our study, we can only say that we have learned a great deal along the way. The image of the river, with its twists and turns and muddy patches, as well as its occasional brilliant vistas, seems exactly right for describing an educational process that has turned out to be even more subtle and complicated than we had imagined it to be when we began our research.


This study is limited in several important respects. First, we are concerned solely with higher education. In our view, one problem with much of the debate over affirmative action is that it lumps together a large number of highly disparate areas and programs, ranging from the awarding of contracts to minority-owned businesses to policies governing hiring and promotion to the admissions policies of colleges and universities. The arguments that pertain to one area may or may not apply in other areas. It is noteworthy, for example, that the plaintiffs in the Piscataway case, which centered on the layoff of a white secondary school teacher, took pains in their final brief to ask the Supreme Court not to confuse the job-specific issues that confronted the plaintiff with the much broader, and rather different, sets of considerations that face educational institutions in deciding whom to admit.

    Within the realm of higher education, we are concerned only with academically selective colleges and universities. The main reason is that the debate surrounding race-sensitive admissions is relevant primarily within these institutions. In colleges and professional schools that admit nearly every qualified applicant, there is little to debate (although there may be arguments over how "qualified" should be defined, and whether the same definition is applied to white and black candidates). It is when there are strict limits on the number of places in an entering class and far more qualified applicants than places, that the choices become difficult and the issue of whether to give weight to race comes to the forefront. Many very well-regarded public universities have broadly inclusive admissions policies at the undergraduate level, and the overall number of selective undergraduate schools is much smaller than many people assume (see Chapter 2). At the graduate and professional level, many schools also take almost every qualified applicant; however, the leading private and public institutions, including almost all accredited schools of law and medicine, select their students from an appreciably larger number of qualified candidates.

    The scope of our study is limited in a third way: although we include information about Hispanic students, our work focuses principally on whites and African Americans (whom we usually refer to as "black"). We hope that other inquiries will be able to do full justice to the educational experiences of Hispanics along with those of Native Americans and Asian Americans. One reason for focusing on black and white students in this study is that so much of the debate over race-sensitive admissions policies has centered on black-white comparisons. There are also practical considerations. While Hispanics share many of the problems faced by blacks, there are so many differences in cultures, backgrounds, and circumstances within the broad Hispanic category that any rigorous study would need to make more distinctions than are possible within the confines of our database. Native Americans have also endured many handicaps and injustices and have benefited from race-sensitive admissions policies. Nevertheless, their representation at the academically selective colleges and universities is exceedingly small and does not permit proper statistical analysis in a study of this kind. Thus, however much we would have liked to include comparisons with a variety of groups of Hispanic and Native American matriculants, this was not a practical possibility.

    Asian Americans differ from other minorities in important respects. Unlike the case of blacks and Hispanics, the percentage of Asian Americans in selective colleges and universities is far higher than their percentage in the population at large and continues to increase at the institutions included in this study. While there are important and sensitive issues associated with the enrollment of Asian American students (who, like Hispanics, are themselves highly diverse), these are different issues from those that confront admissions offices in considering black candidates.

    Finally, our study addresses issues of educational policy. Our objective is not to analyze the development of constitutional law, the proper interpretation of civil rights legislation, or the present holdings of the courts in these areas. We are concerned with the admissions policies that colleges and universities have followed and with their consequences for the country.


Much of the new content in this study derives from exploitation of a rich database called College and Beyond (C&B). This database was built by The Andrew W. Mellon Foundation over nearly four years (from the end of 1994 through 1997) as a part of the Foundation's broader interest in supporting research in higher education. A full explanation of its construction and its components, including links to data compiled by other researchers, is contained in Appendix A. In brief, the part of the database used in this study contains the records of more than eighty thousand undergraduate students who matriculated at twenty-eight academically selective colleges and universities in the fall of 1951, the fall of 1976, and the fall of 1989. Created on the explicit understanding that the Foundation would not release or publish data that identified either individual students or individual schools, it is a "restricted access database."

    The "in-college" component of the database was compiled from individual student records in collaboration with the participating colleges and universities. For each entering student (except those few cases where records had been lost or were incomplete), the database contains information available at the time the student was admitted, including race, gender, test scores, rank in high school class, and, for many students, information about family background. It also includes records of academic performance in college, compiled mainly from transcripts, which have been linked to the admissions data. Each student record was coded to indicate graduation status (when and if the student graduated), major field of study, grade point average, and whether the student participated in athletics or other time-intensive extracurricular activities.

    For many of these same matriculants, we also have extensive survey data describing their subsequent histories (advanced degrees earned, sector of employment, occupation, earned income and family income, involvement in civic activities, marital status and number of children). The respondents were also asked to provide information about where else they applied to college, where they were admitted, whether they did or did not attend their first-choice school, how they now assess their experiences in college, and how satisfied they have been with their lives after college. Finally, for the '89 matriculants only, the survey sought information on the extent to which they interacted (during college and since college) with individuals of different races, political outlooks, socioeconomic backgrounds, and geographic origins. The individuals contacted through the survey were extraordinarily cooperative: the overall sample response rates were 80 percent for the '76 matriculants and 84 percent for the '89 matriculants (Appendix A).

    The twenty-eight colleges and universities whose matriculants are included in the C&B database are:

Liberal Arts Colleges

Barnard College
Bryn Mawr College
Denison University
Hamilton College
Kenyon College
Oberlin College
Smith College
Swarthmore College
Wellesley College
Wesleyan University
Williams College

Research Universities

Columbia University
Duke University
Emory University
Miami University (Ohio)
Northwestern University
Pennsylvania State University
Princeton University
Rice University
Stanford University
Tufts University
Tulane University
University of Michigan at Ann Arbor
University of North Carolina at Chapel Hill
University of Pennsylvania
Vanderbilt University
Washington University
Yale University

Thus the database includes both liberal arts colleges and research universities, including four public universities, and it reflects some reasonable geographic spread. These colleges and universities are not, however, at all representative of American higher education. They were not intended to be. All of them share the attribute of being academically selective, though the degree of selectivity (as measured by the average combined verbal and math SAT score of the entering class) varies considerably.

    In the fall of 1976, eight of the twenty-eight C&B schools had average combined SAT scores of more than 1250 (before the recentering of the scores by ETS which has raised all the scores). Nationally, we estimate that there were only twenty schools in this category, and the eight C&B schools enrolled 40 percent of all freshmen entering these extremely selective colleges and universities. Another thirteen of the C&B schools had average scores of 1150 to 1250; nationally, there were fifty-three schools in this range, and the thirteen C&B schools enrolled 34 percent of all their freshmen. The remaining seven C&B schools had average SAT scores in the 1000-1149 range, and they enrolled 7 percent of all freshmen who entered the 241 schools with SAT scores in this range. In short, the C&B student population contains a sufficiently large fraction of the total number of matriculants at the most selective colleges and universities that we are reasonably confident that our findings apply generally to this set of institutions and especially to those with average scores above 1150.

    In building the C&B database, the intention was to assemble data from a group of schools that were similar enough to permit in-depth comparisons, yet different enough to make such comparisons revealing. Being able to observe the full set of entering students at each of the participating institutions is a great advantage in studying a subject such as race-sensitive admissions. The large size and census-like character of the database, the strong similarities among the institutions in curricula and admissions standards (with many overlapping applications for admission), and the ability to form coherent clusters of institutions (defined by degree of selectivity and type of school) combine to permit a closer, more intensive examination of black-white differences in outcomes than is possible in studies using national samples of individuals from a larger and more diverse array of institutions. We wanted to be able to examine in detail black-white differences among finely classified subgroups of students: men and women, those with lower and higher SATs, those majoring in a variety of fields, those going on to graduate study and those stopping after receipt of the BA, and so on. We believe that "the shape of the river" must be studied at this level of detail if its course is to be charted accurately.

    The other side of the proverbial coin is that because the database was not designed to be "representative," we cannot extrapolate findings from these institutions to the whole of higher education. There are, however, national longitudinal databases that do permit researchers to work with sample data for schools that are much more representative of higher education in general. The objective was to complement the existing longitudinal databases by creating a new resource that would permit more detailed analyses within a circumscribed set of institutions.


This study is highly quantitative. In describing and presenting our work, we have used the simplest techniques that are consistent with the obligation to report meaningful results. Most of the findings are presented in the form of tabulations or cross-tabulations, and we make extensive use of bar charts and other figures (from which the main story line of the book can be read).

    We also use other standard techniques, primarily multivariate regressions, to disentangle the many forces that jointly affect student performance in college, receipt of advanced degrees, and later-life outcomes. While we have no doubt failed to include enough of this finer-grained analysis to satisfy many empirically minded social scientists, we may well have included too much for readers who want only to know "the bottom line." (A considerable amount of explanatory material appears in footnotes.) Our goal has been to achieve the balance that allows us to isolate the effects of different variables--and to understand their interactions--without drifting too far from commonsense questions and answers. Throughout, we have done our best to explain our findings and our methods in language that lay readers can understand.

    The methods used to analyze the data are described in Appendix B. We have also included a great deal of material in additional tables in Appendix D in an effort to make it as easy as possible for readers to check our interpretations, and, if they choose, to substitute their own. In due course, we expect others, using more sophisticated econometric techniques, to extend the analysis presented here. In many instances, the simple methods we employ can only suggest directions and permit what we hope are informed judgments concerning relationships.

    We have devoted a great deal of effort to providing precisely defined national benchmarks that allow the results for the C&B schools to be seen in context. It is important, for example, to compare the earnings of the black graduates of the C&B schools with the earnings of all black holders of BAs who graduated at roughly the same time and to provide the same data for white graduates. In making all such comparisons (as well as comparisons among various groupings of schools included in the C&B database), we confront the problem of selection bias. The process by which students choose colleges and by which colleges choose students is, of course, anything but random, and such a complicated selection process produces outcomes that are independent of the variables we are able to study. We have done our best to deal with this problem by introducing appropriate controls and by attempting to calibrate some of the remaining effects of this double-selection process, but we do not claim to have found a full resolution to this often intractable problem.

    In addition to the many statistics, figures, and tables, we have included in the book some brief personal reflections provided for the most part by individuals who participated in the C&B surveys. These accounts are intended to be only illustrative. Our hope is that they will provide some sense of the kinds of experiences and feelings that underlie the rather antiseptic numbers that appear in such abundance. We would have been reluctant to include these observations--even though many of them are quite revealing--had we not first built the statistical foundation upon which they rest. The stories are meant to amplify the empirical findings and to be thought-provoking, but not to "prove" or confirm any of our interpretations.


Chapter 1 describes the origins and evolution of race-sensitive admissions policies in the context of other changes in American society.

    Chapter 2 discusses the admissions process and describes how race affects the odds of being admitted to selective colleges. The chapter then proposes an operational definition of a "race-neutral" standard and develops estimates of how many black students in the '89 cohort would not have been admitted to certain C&B schools if such a standard had been applied.

    Chapter 3 describes how 1976 and 1989 matriculants fared academically in college--the number who graduated, the majors they chose, how the grades of students varied with their SAT scores, and how black students performed in relation to how we might have expected them to do on the basis of pre-collegiate indicators.

    Chapter 4 follows the '76 and '89 matriculants from college to graduate and professional schools and charts how many of them (classified by rank in class as well as race) went on to earn PhDs or degrees in professional fields such as law, medicine, and business.

    Chapter 5 explores how the 1976 C&B matriculants have done in the marketplace--how many are employed, how much money they have earned, and how satisfied they are with their jobs. We compare blacks and whites, women and men, and C&B graduates with graduates of all colleges nationwide.

    Chapter 6 is concerned with the lives of C&B matriculants outside of the workplace. We examine their civic contributions, marital status, family income, and their own assessments of how satisfied they are with their lives.

    Chapter 7 describes the matriculants' responses when asked to look back and give their impressions of what they learned in college, and whether, given the opportunity, they would go back to the same school, choose the same major, and spend their time in the same ways.

    Chapter 8 examines how much interaction took place across racial and other lines among the 1989 C&B matriculants and reports on the extent to which students from three different eras (those who entered in 1951, 1976, and 1989) agree or disagree with the degree of emphasis that their colleges have placed on recruiting a racially and ethnically diverse student body.

    In Chapter 9, we draw together the major findings from the earlier chapters and discuss their implications for the principal arguments that have been used to criticize race-sensitive admissions policies.

    Finally, in Chapter 10, we present our own conclusions concerning the role of race in the admissions process and how concepts such as "fairness" and "merit" should be interpreted.


There is no way we can thank adequately the small army of people who have worked so hard on this study. The evident importance of the subject, and the privilege of being able, perhaps, to contribute something of value to a wrenching national debate, surely account in large measure for the willingness of all of those mentioned below, and others not mentioned, to go far beyond any definition of the call of duty.

    We begin by thanking our four principal collaborators:

* James Shulman, The Andrew W. Mellon Foundation's Administrative and Financial Officer as well as a program and research associate, has had a hand (and a considerable brain) in every facet of the study. He deserves principal credit for having worked tirelessly with individuals at the institutions in the College and Beyond database, as well as with colleagues at the Foundation and others outside it, to guide the construction of a scholarly resource of immense value. He then participated actively in the analysis of the data, in the drafting and editing of chapters, and in the final passage of the manuscript through what must surely have seemed like an endless swamp rather than a smooth flowing river.

* Thomas Nygren, Director of the Princeton office of the Foundation and also the Foundation's Director of Technology (as well as the program officer responsible for grantmaking in South Africa), has overseen all of the technical work that went into the building of the C&B database with characteristic skill and patience. He has also been responsible for supervising the regression analysis and has taken a principal role (with Stacy Berg Dale) in drafting Appendix A and Appendix B and, more generally, in insisting that the subject deserves the most thoroughgoing effort to respect the underlying data.

* Stacy Berg Dale, a Research Associate in the Foundation's Princeton office, has mastered the intricacies of the C&B database. We have depended heavily on her unusual talent for thinking--by reflex, it seems--in terms of multivariate regressions. She has used this talent to challenge conclusions that might have been accepted at face value, to range freely in exploring alternative hypotheses, and to keep to an absolute minimum the number of arbitrary assumptions and inconsistencies that intrude on any research of this kind.

* Lauren Meserve, a Research Associate in the Foundation's New York office, has worked tirelessly to ensure that the underlying empirical analysis was done correctly and to design the charts and figures that provide the main storyline for the analysis. She has an exceptional range of quantitative and qualitative skills and has been, from start to finish, tenacious in using these talents to improve the research and the presentation of the results.

It is no exaggeration to say that this study could not have been done without the crisp intelligence and unflagging dedication of these four collaborators.

    Other colleagues at the Foundation also made valuable contributions. In the Princeton office, Susan Anderson checked and re-checked the text, made many suggestions for improving the exposition, assisted in the preparation of the list of references cited, and was our principal liaison with Princeton University Press. Douglas Mills was enormously helpful in providing advice on statistical questions and in extracting data from the Census and other national databases. Joyce Pierre, Dorothy Westgate, Jennifer Dicke, and Deborah Peikes all made important contributions to what was clearly a group effort. Earlier in the project, Fredrick Vars, now completing his studies at the Yale Law School, was instrumental in constructing the institutional files that underlie the C&B database and in doing initial empirical work on black-white differences in the relation between SAT scores and academic performance. In New York, David Crook also helped organize data and explore various empirical questions.

    Still other Mellon Foundation staff members provided an unfailing stream of criticism and suggestions as they read versions of the manuscript. Foremost among this group is Harriet Zuckerman, who read more versions of the manuscript than anyone and did so much to improve the clarity of both the analysis and the exposition. Mary Patterson McPherson, T. Dennis Sullivan, Stephanie Bell-Rose, Jackie Looney, and Henry Drewry also read the manuscript carefully and made useful comments. Pat Woodford, Kamla Motihar, and Ulrica Konvalin proved over and over again their willingness to do whatever was needed to bring the project to conclusion. In Cambridge, Connie Higgins has been of enormous help in a project that tested the patience of all who were caught up in its wake.

    We are very fortunate to have benefited from a close reading of the manuscript by outstanding scholars who contributed many valuable suggestions: David Featherman, Director of the Institute of Social Research at the University of Michigan; Randall Kennedy, Professor of Law at the Harvard Law School; Alan Krueger, Bendheim Professor of Economics at Princeton University; three other economists who are now college or university presidents--Richard Levin (Yale University), Michael McPherson (Macalester College), and Harold Shapiro (Princeton University); Michael Nettles, Professor of Education at the University of Michigan; Sarah Turner, Assistant Professor of Education and Economics at the University of Virginia; and Gilbert Whitaker, another economist who is now Dean of the Jones Graduate School of Administration at Rice University. Professors Richard Light of Harvard University, Daniel Kahneman of Princeton University, John Simon of the Yale Law School, and Claire Simon commented on particular chapters. Charles E. Exley, Jr., retired Chairman of NCR Corporation and a Trustee of The Andrew W. Mellon Foundation, also read the manuscript with great care and made extremely insightful comments--which he transmitted to us from the Sudan! At an earlier stage in the study, Professors David Card, now at the University of California at Berkeley, and Orley Ashenfelter of Princeton University, contributed to the shaping of the research design.

    Robert K. Merton of Columbia University and Arnold Rampersad of Princeton University provided knowing advice concerning the title of the book and the preface. Alan Rosenbaum, Director of the Art Museum at Princeton, was heroic in his efforts to find just the right cover illustration (he succeeded, we think).

    In our initial efforts to collect institutional records, we were joined by an exceptional group of people at the twenty-eight participating institutions, many of whom worked nights and weekends to generate the raw files we needed. It is only limitations of space that prevent us from thanking each of them, and their presidents, for having had the faith to participate so actively in the construction of the C&B database.

    The survey component of the database, which plays such a vital role in the analysis, could not have been created without the thoughtful contributions of Herbert Abelson of the Survey Research Center of Princeton University and Geraldine Mooney and her colleagues at Mathematica Policy Research (the entity that administered the surveys so successfully, as is documented in Appendix A). We also want to thank the forty-five thousand individuals who took the time to complete the surveys so carefully and often volunteered additional comments. Many of these former students obviously care, and care deeply, about the questions we have been studying.

    As a companion project, the Foundation commissioned the creation of a national control group survey (described in Appendix A); Norman Bradburn and Allen Sanderson of the National Opinion Research Center in Chicago did yeoman work in completing this part of the project.

    We were also able to link the core of the C&B database to two other large databases that complemented the information we were able to collect directly. Donald Stewart and his colleagues at the College Entrance Examination Board and the Educational Testing Service, and Alexander Astin, Director of the Higher Education Research Institute at UCLA, and his colleagues at the Cooperative Institutional Research Project, understood what we were trying to accomplish and were determined to help. In addition, Linda Wightman, former Vice President of the Law School Admission Council and now a faculty member at the University of North Carolina, Greensboro, went to extra efforts to provide detailed data on law school students.

    In order to learn more at first hand about the interest of businesses and professional associations in the recruitment of minority students, we contacted many knowledgeable individuals. Thomas Schick at American Express, Ira Millstein and Marsha Simms at Weil Gotschal & Manges, Jeffrey Brinck and Christina Wagner at Milbank Tweed Hadley & McCloy, Richard Fisher and Marilyn Booker at Morgan Stanley, and Marc Lackritz at the Securities Industry Association in Washington, D.C., were all extremely generous with their time. Subsequently, Glenda Burkhart has been responsible for involving representatives of the business, professional, and academic communities in thinking about the implications of this research.

    We have been fortunate, too, in our publisher. Walter Lippincott, Peter Dougherty, Neil Litt, and their colleagues at Princeton University Press made it clear from the outset that for them this project was in no way "business as usual." They have worked diligently to publish a complex book at their usual high standard under extraordinary time constraints.

    Finally, we wish to thank the Trustees of The Andrew W. Mellon Foundation for their appreciation of what we have tried to do, their financial support, and their understanding (nay, their insistence) that we would, of course, come to our own conclusions. The arguments developed in this book represent our own thinking, and none of the Trustees of the Foundation, nor any of the others who provided so much advice and help, should be implicated in the results. Whatever faults remain, despite the efforts of so many to "get it right," are solely our responsibility.

William G. Bowen
Derek Bok
May 1998

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
Princeton Legacy Library
Exam/Desk Copy
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share