Skip to main content

Using Structural Equation Modeling in Community College Research: Comparing the Effects of Psychological Latent Factors Between Hispanic and Non-Hispanic Students

Case
By: Published: 2017 | Product: SAGE Research Methods Cases Part 2
+- LessMore information
Search form
No results
Not Found
Download Case PDF

Abstract

This case presents an empirical study that examined how general self-efficacy functions as a predictor of community college student’s degree aspiration directly and indirectly via various key factors (i.e., social capital, personal input, transfer readiness). This study is unique because the author (a) utilized general self-efficacy scale and (b) uncovered variations between Hispanic and non-Hispanic students in terms of how general self-efficacy took effects.

Structural equation modeling techniques were adopted in this study. In particular, the author first developed the hypothetical general self-efficacy model for community college students through literature review. Next, by analyzing large-scale survey data, a confirmatory factor analysis confirmed a measurement model for the general self-efficacy model for community college students. Then, the structural equation modeling analysis tested the tenability of general self-efficacy model for community college students among Hispanic and non-Hispanic students in the sample. Finally, a multi-group invariance testing revealed significant model differences between Hispanic and non-Hispanic student groups.

This research case focuses on discussing the process of structural equation modeling analysis. For example, the author describes the procedures of constructing the hypothetical model, testing the model, finalizing the model, and conducting the multi-group invariance tests. Challenges of finalizing the measurements in confirmatory factor analysis and modifying the general self-efficacy model for community college students are discussed. In addition, this case also discusses the rationale of using or not using alternative statistical technologies. In particular, the author describes the decision-making process of not using statistical matching techniques (such as propensity score matching) to create comparison groups in multi-group invariance testing.

Looks like you do not have access to this content.

Methods Map

Structural equation modelling

Copy and paste the following HTML into your website