The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions.

• Author: Bates, Timothy
• Position: Review

By William G. Bowen and Derek Bok.

Princeton, NJ: Princeton University Press, 1999. Pp. xxxvi, 472. $24.95.

The Shape of the River is a provocative, genuinely fascinating book. Doctors Bowen and Bok, past presidents of Princeton and Harvard, respectively, have overseen the collection and analysis of massive time-series databases on the lifetime achievements of tens of thousands of college students, and they have produced a landmark work. That the book has some rather pronounced shortcomings is not a serious deficiency because the authors have, after all, tackled some very difficult empirical problems. Knowledgeable econometricians will certainly note that the strong conclusions offered by Bowen and Bok sometimes outrun what their analyses are capable of supporting. Yet the straightforward econometric models that make up the core of the book's analytical content are creative and the outcomes are thought-provoking. This volume moves forward our understanding of the current debate about the wisdom of race-sensitive college admissions, and it will stand as a most important and influential study of affirmative action generally in the forseeable future.

Amid all of the facts and figures, one conclusion stands out: The nation's most selective colleges and universities have succeeded both in increasing substantially their enrollment of minority students in recent decades and in educating sizable numbers who have already gone on to achieve considerable success (however measured). Rest assured that numerous alternative measures of success were explored. In light of all the answers provided in the Bowen--Bok study, it is useful to ask oneself repeatedly, while reading this volume, what the questions are?

The authors seek to measure the consequences of race-sensitive admissions policies that the nation's most selective schools have pursued now for 30-plus years. No matter how many alums one chooses to survey, one is still left with a well-known phenomenon called selection bias. Much of this book is devoted to reporting the results of a statistical study of 1951, 1976, and 1989 matriculants at 28 highly selective colleges and universities. Some 45,000 of them responded to the survey, but I would really like to know what would have happened to those respondents, particularly the minority students, if affirmative action had never been initiated. A random assignment experiment would have been nice. Harvard, for example, could have selected 600 minority...