Estimating the Causal Effect of Learning Approaches on College Students' Four-Year Graduation

• In Event: Enhancing First-Year Student Success: Insights From Programs and Courses

Abstract

This research focuses on using the 2017 NSSE (National Survey of Student Engagement) data collected from a large, public university to estimate the causal effects of different learning approaches students employed during their freshman year, on their 4-year graduation. Four scales (Reflective, Integrative, and Higher-Order Learning; and Learning Strategies) were selected from NSSE and generated a 4-class model using the “Bolck, Croon, and Hagenaars (BCH)” approach to latent class analysis (LCA) with distal outcomes. After incorporating the inverse propensity score weights as sample weights in the outcome analysis, the current study found that freshmen who reported to be Very frequent strategy users and Frequent strategy users had higher probability of timely graduation.

Authors

• Siyu Liu, University of South Florida

• Amber D. Dumford, University of South Florida - Tampa

• Eunsook Kim, University of South Florida