Former UC President Richard Atkinson Reviews "Crossing the Finish Line"

Science 11 September 2009:
Vol. 325. no. 5946, pp. 1343 - 1344
DOI: 10.1126/science.1179882




Addressing the Graduation Gap

Richard C. Atkinson1 and Saul Geiser2

Crossing the Finish Line Completing College at America's Public Universities by William G. Bowen, Matthew M. Chingos, and Michael S. McPherson Princeton University Press, Princeton, NJ, 2009. 413 pp. $27.95, £19.95. ISBN 9780691137483.

1 5320 Atkinson Hall, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093–0436, USA.
2 Center for Studies in Higher Education, University of California, Berkeley, CA 94720–4650, USA.



E-mail: sgeiser@berkeley.edu

Since retiring as president of Princeton and assuming leadership of the Mellon Foundation, William G. Bowen has been lead author of an extraordinary series of books on topics in American higher education such as affirmative action (1), the role of intercollegiate athletics (2), and access for low-income students (3). Bowen's hallmark has been "building large, linked databases and seeing what lessons can be learned from them." Crossing the Finish Line, written with Matthew M. Chingos, a Mellon associate, and Michael S. McPherson, president of the Spencer Foundation, continues in this tradition but on an even grander scale.

The book is concerned with the alarming slowdown of improvement on most measures of educational attainment, particularly college completion rates, since the 1970s. During the first three-quarters of the 20th century, the United States made substantial progress in improving both access to higher education and completion of baccalaureate degrees. That progress slowed dramatically in the mid-1970s and has since remained almost flat. With its current college-completion rate of 56%, the United States now ranks near the bottom of the 30 Organisation for Economic Co-operation and Development countries. And among its college-age population, disparities in college graduation by race, ethnicity, and socioeconomic status have continued and in some cases worsened.

Whereas Bowen's earlier books focused on elite, private institutions, public institutions account for over three-fourths of college enrollments in the United States. For Crossing the Finish Line, he and his colleagues constructed a massive, longitudinal database of students at 21 public "flagship" universities and four state university systems. Students were tracked for six years following their matriculation in 1999. The unprecedented scale of the database and its linkage to student-level, standardized-test data permit the most comprehensive look yet possible at the determinants of graduation rates—and what might be done to improve them.




CREDIT: JOHN MCCORMICK/VIRGINIA TECH

[Larger version of this image]

Bowen and McPherson are economists and bring economists' sensibilities and methods to their subject. Much of the book uses regression analysis to assess the impact of various factors on college completion (e.g., socioeconomic status, financial aid, and institutional selectivity) after adjusting for other factors such as students' high-school grades and test scores. Individual chapters deftly summarize what is known about each topic and then often extend that knowledge substantially.

One finding likely to draw wide attention is that, of the criteria commonly employed in college admission decisions, high-school grades are by far the best predictor of college completion. Once those grades are known, scores from the SAT and the ACT add little to the prediction. This pattern holds irrespective of the quality or type of high schools that students attend and across colleges and universities at all levels of selectivity. The authors also find that grades are less closely associated with socioeconomic status than are SAT or ACT scores. Together these findings suggest that selective institutions could admit students from a broader range of socioeconomic backgrounds, with no deterioration in graduation rates, by giving greater weight to high-school grades.

Yet the authors by no means close the door on standardized admissions tests. They find that achievement tests (such as the SAT Subject Tests and Advanced Placement exams) that measure students' knowledge of specific college preparatory subjects are better predictors of college completion than the generic SAT or ACT exams. Subject-based admissions tests also have important "signaling" effects for high schools, encouraging teaching and learning of a more rigorous academic curriculum. Crossing the Finish Line will surely enliven and deepen the national dialogue about the role of standardized tests in college admissions.

Another provocative finding concerns "undermatching": students enrolling in colleges that are less demanding than they are qualified to attend. Using a conservative standard—grades and test scores that would qualify students in the top 10% of admits at highly selective public institutions—the authors find that a substantial proportion of well-qualified students enroll instead at two-year or less-selective four-year institutions. The pattern is most pronounced among low-income and minority students. Counterintuitively, however, undermatching has a negative effect on graduation rates. College completion varies sharply with institutional selectivity, even after controlling for student characteristics. As a result, well-qualified students who attend less-selective colleges graduate at lower rates than those with comparable qualifications at highly selective institutions. As the authors conclude, "The scale of the undermatch phenomenon among students from modest backgrounds suggests that addressing this problem offers a real opportunity to increase social mobility and simultaneously to increase overall levels of educational attainment."

Recently there has been a revival of interest in expanding transfer from two-year to four-year institutions as a means of raising college completion rates, and Bowen, Chingos, and McPherson bring their massive database to bear on that issue as well. On one hand, they find that at many four-year institutions, students who transfer from a community college are more likely to earn a bachelor's degree than students who begin as freshmen, suggesting that those institutions might improve completion rates by admitting more transfers. On the other hand, given the considerable attrition within two-year institutions, enrolling there can negatively affect a student's chances of earning a four-year degree. The authors find that highschool seniors who enter two-year colleges seeking a bachelor's degree are much less likely to complete their degrees than comparably qualified students who move directly from high school into a four-year program. One of the book's strengths is the authors' willingness to follow wherever the evidence leads, even where, as in this instance, the findings may seem to conflict.

The book provides new and often surprising insights on other major determinants of college completion. The chapters on financial aid, in particular, are masterful. The authors argue that, for public universities, offering more need-based student aid "is demonstrably less expensive than keeping the net price low by reducing tuition across the board—a policy that provides further subsidies to well-off families without improving their graduation rates."

Crossing the Finish Line also breaks new ground in its manner of presentation. The thread of the analysis can be read directly from the figures interspersed liberally throughout the text. Complex regression analyses are reduced to a graphic language that helps propel the reader through the argument. And in another innovation one hopes others will emulate, the book features an online appendix, maintained by Princeton University Press, where readers can explore a vast array of supporting data for each chapter to judge for themselves whether its conclusions are warranted.

Why have college completion rates in the United States stagnated since the 1970s? Databases of the kind Bowen and his colleagues have built do not necessarily lend themselves to historical analysis, and the authors do not attempt a complete answer to this question. Yet a clue may be found in an essay by another Mellon associate, Eugene M. Tobin, presented as an appendix to the book. Tobin traces the modern history of state public university systems after World War II. Many state systems, he finds, were influenced by the California model and its tripartite structure of highly selective research universities, comprehensive four-year institutions, and open-access community colleges. But as that model evolved during the latter part of the 20th century, most of the growth in student enrollment has been absorbed by the community colleges and four-year comprehensives. Undergraduate enrollments at the flagship campuses generally have not kept pace. Given the strong association between institutional selectivity and college completion, the changing structure of state higher education systems may be an important piece of the puzzle.

Crossing the Finish Line exemplifies the best that social science research has to offer: rigorous empirical analysis brought to bear on a major public policy issue. Bowen, Chingos, and McPherson have provided an essential resource that both researchers and policy-makers will consult for years to come.

References

1. W. G. Bowen, D. Bok, The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions (Princeton Univ. Press, Princeton, NJ, 1998); reviewed in (4).
2. W. G. Bowen, S. A. Levin, Reclaiming the Game: College Sports and Educational Values (Princeton Univ. Press, Princeton, NJ, 2003).
3. W. G. Bowen, M. A. Kurzweil, E. M. Tobin, Equity and Excellence in American Higher Education (Princeton Univ. Press, Princeton, NJ, 2005).
4. R. E. Thach, Science 284, 1473 (1999).[Free Full Text]