Comparing Meta-Analytic Structural Equation Modeling Approaches across Model Assumptions Using an Empirical Example

Meta-analytic Structural Equation Modeling (MASEM) is the combination of meta-analysis (MA) and structural equation modeling (SEM). With new MASEM methodologies developed over the past few years, there is an opportunity to compare the past approaches with the new ones. The purpose of this dissertation is two-fold. First, the parameter estimates, standard errors, confidence intervals, and heterogeneity measures are compared across 8 MASEM approaches (fixed-effect and random-effects univariate r, fixed-effect and random-effects univariate z, fixed-effect and random-effects Two-Stage SEM approach, and fixed-effect and random-effects One-Stage MASEM approach) using 25 studies relating to college persistence. Overall, results found only slight differences in estimates across methods (differences to two or three decimal places). The biggest difference was found in significant path estimates between univariate and multivariate approaches, which is primarily due to sample size differences. The second purpose of this paper was to synthesize the current literature pertaining the relationships between student background characteristics, institutional characteristics, academic integration, and social integration on student success. Results indicate that student background characteristics, academic integration, and social integration had a direct significant impact on student success. Although institutional characteristics did not have a significant impact on student success directly, it had a significant impact on academic and social integration. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]