Abstract
This research investigates institutional-level retention rates at 784 community colleges in the United States over a 16-year period (2004–2019). We analyse institution-level panel data from the Integrated Postsecondary Education Data System (IPEDS) and employ a fixed-effects method to estimate how institutional-level retention rates relate to the time-varying student-body characteristics of 15 entering cohorts, time-varying institutional-level characteristics and actions and economic conditions. Consistent with previous research, we consider the student body as a whole (full- and part-time students combined) and find that student-body demographic composition and financial assistance are associated with institutional-level retention rates. Institutional time-varying faculty composition and offerings, as well as institutional spending and tuition, are also linked to retention rates. County-level market conditions drive retention rates. Following further statistical tests, we extend our institutional-level analysis to the full- and part-time student bodies separately to identify factors linked to improved retention for each. Our results identify inconsistencies between the aggregate student-body analysis (the standard in the literature) and the individual full-/part-time analyses. Specifically, some results that are significant in the aggregate sample are no longer significant in the individual studies. Other estimates have opposite signs for the full- and part-time population. Furthermore, some time-varying characteristics are statistically significant in the individual population studies, but not in the aggregate analysis. These findings open the conversation about the best way to study community college institutional-level retention rates to inform both campus managers and policymakers in their strategic, financial and managerial decision-making processes.