Many colleges and universities, both public and private, provide grant aid to undergraduates to help them pay for all or part of the tuition and fees charged by the institution. This practice, often referred to as “tuition discounting,” has grown rapidly in recent years (Redd 2000; Cunningham et al. 2001; Hubbell and Lapovsky 2002). Depending on the type and selectivity of the institution, institutional aid is awarded for different reasons. Some institutions aim to promote access to low-income and otherwise disadvantaged students, others use institutional aid to increase the enrollment of meritorious students, and still others use it to increase tuition revenues (Allan 1999; Redd 2000). Many institutions are trying to accomplish more than one of these goals simultaneously (Redd 2000). Through the packaging of need-based and merit-based aid, different institutions use different strategies. For example, a need-within-merit strategy uses merit criteria, but prioritizes the recipients on the basis of need, whereas a merit-within-need strategy awards aid on the basis of need, but prioritizes the recipients on the basis of merit.

This study provides information about recent trends in institutional aid receipt and then examines the relationship between such aid and the likelihood of recipients staying enrolled in the awarding institution relative to comparable unaided students. The trend analysis is based on data gathered from three administrations of the National Postsecondary Student Aid Study, conducted in 1992–93, 1995–96, and 1999–2000 (NPSAS:93, NPSAS:96, and NPSAS:2000), and the retention analysis is based on data from the first and second follow-ups to the 1995–96 Beginning Postsecondary Students Longitudinal Study (BPS:96/01). BPS followed a cohort of students who first enrolled in college in 1995–96 and were last surveyed in 2001, about 6 years after their initial enrollment. Only full-time students attending 4-year public and private not-for-profit institutions were included in these analyses.

USER NOTE: This publication is best viewed using a screen resolution of at least 800x600 pixels. For instructions on how to change your screen resolution, please see NCES Help.