Summer Engagement and Student Persistence at a Selective Private University

• In Event: 73.047 - The Not-So-Lazy Days of Summer: Interventions to Improve Achievement and Attainment

Abstract

Postsecondary education is an increasingly important determinant of long-run socioeconomic outcomes and upward socioeconomic mobility (e.g., Ellwood & Kane, 2000). It is troubling, then, that differences by gender and socioeconomic status (SES) in college attendance, persistence, and completion rates have grown over the past thirty years (Bailey & Dynarski, 2011). These trends have received attention from policymakers, educators, scholars, and pundits. However, the underlying causes of these phenomena remain unclear, which limits the ability of policymakers and educators to devise policies that provide equal access to the educational opportunities necessary for all students to experience happy, fulfilled, and successful lives.

Heyns (1978) first suggested that differential rates of summer learning by SES may contribute to the persistence of racial and socioeconomic achievement gaps. Several studies have since documented significant differences between SES groups in the development of reading and literacy skills during summer vacation (e.g., Alexander, Entwisle, & Olson, 2001; Burkam et al., 2004; Downey et al., 2004). However, the existing literature on summer learning focuses almost entirely on K-12 schooling, despite the likelihood that similar phenomena affect postsecondary persistence and completion (Castleman & Page, 2014). The proposed research will begin to fill this gap in the literature by addressing the following research questions using longitudinal administrative student-by-course “big data” from a selective, not-for-profit, private, Mid-Atlantic university:

1) What observable characteristics predict the type and amount of postsecondary students’ summer engagement?

2) Are students who remain engaged with the university over the summer months – by taking classes on campus, taking classes online, or working in some capacity on campus – more likely than those who do not remain engaged to persist toward completion?

3) Are there heterogeneous effects of summer engagement on persistence and completion?

Specifically, we will address these questions by applying multivariate regression and matching estimators to student-by-semester longitudinal administrative data from American University (AU), which is located in Washington, D.C. and enrolls about 6,000 undergraduate students. These data include information on all courses taken by students who enrolled at AU between the academic years 2000 and 2013 and a wealth of student background characteristics. Preliminary results suggest that there is significant variation in summer engagement, as measured by summer course taking: 8% of rising sophomores enrolled in at least one summer-semester class during the sample’s time frame. Of these students, 95% persisted into the fall semester of their sophomore year, compared to an 88.5% persistence rate for students who did not earn credits during the summer between their freshman and sophomore years. Both male and black rising sophomores enrolled in summer courses at slightly higher rates than did female and white rising sophomores, respectively. As in the overall sample, both of these demographic groups were significantly more likely to persist into the fall semester of their sophomore year after earning credits during the intervening summer. Subsequent analyses will refine these simple tabulations by conditioning and matching on student background characteristics and type of summer coursework (e.g., second course attempts; online courses).

Authors

• Scott Bass, American University

• Seth Gershenson, American University

• David Pitts, American University