Nuancing College Student Pathways With Multiple-Group SEM (Structural Equation Modeling): An Illustrative Example From Computing Graduate Education

• In Event: Measurement Methods for Complex Problems in Higher Education

Abstract

Traditional quantitative methods rely on aggregation, often representing majority student experiences as the norm. However, college student pathways are more nuanced than such methods suggest. The purpose of this paper is to offer an empirical illustration of the potential for multiple-group structural equation modeling (MG-SEM) to serve as a critical quantitative analytical tool by empirically testing heterogeneity in students’collegiate pathways. We demonstrate the utility of MG-SEM by comparing the analytical process and results to those from a traditional SEM approach in the context of computing graduate education. Results indicated that the MG-SEM approach, which modeled distinct pathways along lines of intersecting racial/ethnic and gender identities, more accurately represented the data. Implications of this finding for equity-minded practice are discussed.

Authors

• Christa E. Winkler, Mississippi State University

• Annie M. Wofford, Florida State University

• Linda DeAngelo, University of Pittsburgh