Prediction of persistence at a commuter institution of higher education: a multivariate analysis of a theoretical model

Abstract

Changing demographics suggest that much of the growth in higher education will come from diverse students consuming higher education in nontraditional formats. The majority of investigations in persistence research have focused on traditional students attending traditional, residential institutions. Additionally, Horn and Berger (2004) indicated that although student enrollments have increased over the past two decades, persistence rates have remained unchanged. The purpose of this research was to develop and test a longitudinal model synthesized from the literature to (a) investigate the associations on persistence for nontraditional students attending nontraditional institutions, (b) assist staff, faculty, and administrators implement high quality intervention strategies, and (c) refocus institutional resources toward enhancing student persistence. Data were collected in a three-step process over a 5-year period; 1999 (Year 1); 2000 (Year 2), and 2004 (Year 5). The Year 1 student cohort initially consisted of 4,571; 697 of them also were assessed in 2000, their second year. Results yielded significant differences between students who persisted to their second year or not, those who graduated or not within 5 years, age groups, and ethnicities. Further analyses indicated that satisfaction significantly decreased on every construct in the theoretical model as students moved from their first to second year. Path analysis was employed to test the theoretical model. Two models, one using data from the students' first year of college and one representing the students' second year of college, were assessed. The R2 for intent to re-enroll was .59 for Year 1 and .66 for Year 2. Prior research studying commuter institutions showed background characteristics of the student as the most important associations on short-term outcomes and persistence decisions. Results of this study indicated the opposite; institutional and interactional variables were associated with persistence decisions for commuter students. Furthermore, both academic and social variables showed direct associations on satisfaction and indirect associations on intent to re-enroll. Logistic regressions supported other findings of this study: strong academic performance, high levels of satisfaction, and full-time status each enhanced persistence rates. The findings of this dissertation suggested a revised theoretical model to promote additional quantitative research for this line of inquiry.