Assessing Factorial Invariance in the Presence of Latent Distribution Heterogeneity and Unbalanced Sample Sizes

• In Event: Sunday Roundtable Session 11:25 am
  In Roundtable Session: Measurement Issues in Structural Equation Modeling (Table 13)

Abstract

Factorial invariance is essential for valid comparisons of latent constructs across groups. This study aimed to investigate the performance of various fit measures in testing factorial invariance when the combination of unequal latent factor distributions and unbalanced group sizes was present. We manipulated design factors involving different latent distributions, sample sizes, and levels of non-invariance to examine the false positive rates and sensitivity of each fit measure. The results suggested varying impacts of latent distribution heterogeneity and unbalanced samples on metric and scalar invariance testing. This study enhances the understanding of factorial invariance and improves the generalizability of findings in diverse research settings.

Authors

• Xinya Liang, University of Arkansas

• jiaying chen, University of Arkansas