Taking Aim at the Supply Side
The higher education world often attributes the last few decades' decline in the U.S. college completion rate on high schools, arguing that students are unable to make it through college because they’re inadequately prepared to set foot on campus as freshmen.
A new study, though, pins the bulk of the decline on the change in the types of institutions students are attending. In “Why Have College Completion Rates Declined? An Analysis of Changing Student Preparation and Collegiate Resources,” published Monday by the National Bureau of Economic Research, three economists outline findings suggesting that the shift of a greater proportion of first-time college students to community colleges and non-flagship publics, as well as the declining per-student resources of those institutions, have driven down completion rates.
"The conventional wisdom that it’s all about declining student preparation is not correct,” said Sarah E. Turner, one of the paper’s authors, a professor of education and economics at the University of Virginia. “The evidence is not just about changes in student characteristics but the kinds of institutions larger numbers of students are attending.”
Using data from the National Center for Education Statistics' longitudinal surveys, Turner and her co-authors -- John Bound, a professor of economics at the University of Michigan, and Michael Lovenheim, an assistant professor of policy analysis and management at Cornell University -- found that the proportion of high school seniors enrolling in college within two years of graduating high school rose from 48.5 percent of 1972 high school graduates to 70.7 percent of 1992 graduates. But, among the same group, eight-year college completion rates fell from 50.5 percent to 45.9 percent.
The primary factor effecting the change, the authors argue, has been the rise of the community college as students' first stop toward possibly attaining a bachelor’s degree. In the 1972 cohort, 31.2 percent of students started out at a community college; by 1992, 43.7 percent of students enrolling in college started out at a two-year institution. Eight-year bachelor’s attainment was at 20.2 percent for the 1972 group and fell to 17.6 percent for the 1992 group. Community colleges and non-flagships have higher student-faculty ratios and lower median expenditures per student than flagships and privates, and resources per student fell at both kinds of institutions from the time that the 1972 group was enrolled to when the 1992 group was enrolled.
Their finding: “the supply side of higher education plays an important role in explaining changes in student outcomes …. Our analysis suggests that, at least for changing completion rates, student preparation is only a partial explanation; characteristics of the supply-side of the market have a substantial influence on student success in college.”
Clifford Adelman, a senior associate at the Institution for Higher Education Policy and former Department of Education researcher, said he sees one major problem with the study: “the failure of literature review.” Plenty of other studies have used the same NCES data to study degree completion rates, he said, but those studies weren't taken into account in designing and implementing this study. (Turner said it was "an oversight" not to cite studies conducted by Adelman and others in her paper.)
To Adelman, the researchers' findings are obvious. “Your grandma could have told you that” students who start out at a community college -- controlling for level of academic preparation -- would be less likely to graduate than those who first enroll at a flagship or private institution.
Among the criticisms he listed in an e-mail message to Inside Higher Ed: the authors “confine the population to those who entered degree-granting institutions of higher education within two years of scheduled high school graduation, censor the period across which this population is followed to eight years, concluded that ‘graduation rates’ have fallen, and attributed the lion’s share of causality of the fall to the increased proportion of students from the class of 1992 who started out in community colleges (as well as differential completion rate trends by selectivity of four-year colleges).”
Turner defended the work she and her coauthors produced.”We’ve been pretty careful in the choices that we’ve made and I think our results are really quite robust,” she said.