Panel:
Dual Enrollment and Other Approaches to College Access and Success
(Education)
*Names in bold indicate Presenter
Our panel explores the effects of early college programs in high school on college access and success. The three papers evaluate if programs designed to provide rigorous college coursework to students in high schools in Worcester and New York City increase the likelihood of college readiness and enrollment. The results have policy implications for school districts seeking to increase postsecondary success for all students.
The first paper examines how the enrollment in and completion of AP courses impacts students’ plans after high school graduation. Using data from Worcester County, Massachusetts, the authors analyze a college readiness metric called “post-graduation indicator” for graduating cohorts of high school seniors from 2011 to 2017 across 7 high schools. The indicator is a ranked scale coded from exit surveys in which students indicate their postsecondary plans. Employed as a dependent variable, they run ordered probit regressions and control for ethnic and economic specifics, with two explanatory variables: 1) total number of AP courses taken and 2) an AP access indicator score ranging from 1 to 5. Preliminary results show the number of AP courses as a significant predictor of increases in the post-graduation indicator score, suggesting that each additional AP course raised the possibility that a student would aim to enroll in 4-year colleges.
The second paper explores how school and student characteristics of participants in the Early College Initiative (ECI), a dual enrollment program that partners high schools with specific colleges, impacts the likelihood of college enrollment. Making use of longitudinal data from City University of New York’s (CUNY) Administrative Data Warehouse and the National Student Clearinghouse (NSC), the authors examine the predictive impact of students’ individual- and school-level characteristics on timely college matriculation for the 6,800 students enrolled in an ECI high school from 2006-2017. A binary logistic regression demonstrates that demographic factors such as race, gender, and socio-economic status, while important, may be less predictive when accounting for academic and behavioral factors. Notably, when accounting for early college GPA, early college course withdrawals, and other academic factors, gender is not a significant predictor of college enrollment in the ECI context.
The third study measures the impact of dual enrollment, in high school and college courses, through the College Now (CN) program on early postsecondary outcomes for students in New York City. The sample consists of students from the 74% (443 out of 599) of public high schools where the program is offered from 2014-2016. Using administrative data from NYC and NSC, the authors use both a sharp and a fuzzy regression discontinuity design (RDD) methodology that estimates the local average effect of the treatment, dual enrollment in high school and college classes, on first semester and first-year postsecondary outcomes including college enrollment, number of course credits completed, and freshman GPA. To measure the causal effect, we compare first semester and first-year postsecondary outcomes for students both eligible for and enrolled in CN who fall directly above and those who are not enrolled in CN and are directly below the score cutoff.
