Untangling the Web

Abstract

Fifteen years ago, experts predicted that as a result of the shrinking traditional college-age cohort higher education enrollments would decline in the 1980s and early 1990s. During the 1980s tuition charges consistently increased faster than inflation, state support of higher education wavered because of troubled state economies, and the federal government shifted its emphasis from grants to loans and decreased its support for student aid. In spite of these developments, all of which would have led to more dire enrollment predictions if they had been known a decade earlier, enrollments in higher education remained stable in the 1980s [16], instead of declining as had been predicted. In the 1980s college enrollment was higher than predicted. However, minority participation rates declined, which suggests the decline in federal student aid could have had an influence on enrollment. Throughout the decade there was speculation that the declines in minority participation rates were attributable to changes in federal student aid in combination with rising college prices [for example 36]. However, the linkage between changes in federal student aid policy and enrollment had been seriously questioned [18], and the Reagan administration's studies of the issue concluded that academic preparation, not student aid, was the major reason for the decline in minority participation rates [9, 39]. Recent research, however, supports the argument that there is a An earlier version of this article was presented at the Eighth Annual NASSGP/NCHELP Research Network Conference, San Francisco, Calif., 21 March 1991. The thought-provoking comments of unknown reviewers are gratefully acknowledged. linkage between federal student aid, tuition charges, and enrollment by minorities from low-income backgrounds [31, 45, 47, 48, 51]. Unfortunately, this linkage is not routinely considered in most state higher education agencies or the United States Department of Education when they make decisions about higher education budgets. In spite of the importance of this topic, there has not been any significant effort to collect the data researchers would need to address these concerns fully. Therefore, the adequacy of existing price-response measures for enrollment and financial planning should be assessed. To be useful in budget and enrollment projection models, price-response measures should meet at least two minimum standards. First, they should be at least as accurate in predicting enrollment as projection models that do not use these measures. Second, they should predict the types of changes that actually happen (for example, redistribution of aid and enrollment across sectors). In this article I attempt to untangle the web of controversy surrounding the use of price-response measures in enrollment projection models. After briefly considering the use of standardized student price-response coefficients (SPRCs) and other cashflow-response measures(1) in projection models, I compare the accuracy of three types of cashflow-response measures in predicting enrollment, use one of these to analyze the impact of pricing changes during the 1980s, and consider the implications of these analyses for future refinements in enrollment projections and data collections. The Use of Price Response in Projection Models In the 1960s the early student-demand studies concluded that students' enrollment decisions were influenced by tuition charges [for example, 6, 11]. In 1973 the National Commission on the Financing of Postsecondary Education (NCFPE) used the results of the student-demand studies in a higher education budget model. However, the NCFPE's budget model was severely criticized [12] largely because of inadequacies of the student-demand studies used in the model. Nevertheless, other research on students who enrolled before the Pell program was implemented (as Basic Educational Opportunity Grants), suggested that results of research on the impact of student aid could be useful in estimating the enrollment effects of student aid policy [29]. …