Association for Public Policy Analysis & Management

College access has increased across socio-demographic groups over time, however the gap in degree attainment continues to widen (e.g. Ziol-Guest & Lee, 2016). There is an urgent need to design institutional and programmatic interventions that support students to degree completion. Yet, knowledge about the effectiveness of interventions and the mechanisms that drive and explain college persistence remains underdeveloped. A review of causal research to date suggests that comprehensive programs that offer both financial and social supports yield large impacts on college persistence and completion. For example, a randomized study of the City University of New York’s Accelerated Study in Associate Programs found that the program doubled associate’s degree completion rates (Scrivener et al., 2015). In addition, Clotfelter, Hemelt and Ladd (2017) used a combined difference-in-differences and regression discontinuity design to examine the impact of the Carolina Covenant, and found that the program increased four-year graduation rates by approximately 8 percentage points.

We contribute to this critical research area with an examination of the Dell Scholars Program, a college success initiative that provides selected students with a generous scholarship and ongoing support throughout college. The Dell Scholars Program selects 300 scholars annually from a large, national applicant pool. The program’s recruitment efforts specifically target low-income students who maintain at least a 2.4 GPA in high school and plan to enroll full-time in a four-year college. Compared to certain other merit scholarship programs, such as the Gates Millennium Scholars, this program is unique in that it serves students attending colleges that vary across key institutional characteristics, such as selectivity. The program model warrants deep investigation for several reasons: their use of a web-based system to collect extensive, validated information on student progress; their proactive approach to providing social support in addition to financial assistance; their ability to support students efficiently with a small project team; and the sizable observed effects on degree attainment outcomes.

We will present both quantitative and qualitative findings on the programs impact. First, we will summarize results from a quasi-experimental study of the program’s impact on college persistence and degree completion (Page, Kehoe, Castleman & Sahadewo, forthcoming). For example, we find that the program leads to substantially higher rates of bachelor’s degree completion within six years as well as improvements on multiple other measures of college success. Completion impacts are particularly large for students attending less-selective institutions where institutional supports may be less robust. We will build on these causal estimates by presenting findings from student surveys and student and staff interviews to shed light on how the program operates to engender these outcomes. We will discuss implications for research, policy and practice.