Detecting Latent Classes Through Mediation in Regression Mixture Models

• In Event: Fancy Structural Equation Modeling and Hierarchical Linear Modeling

Abstract

The current study aims to investigate latent class enumeration accuracy with mediation in regression mixture models. Little research has examined this technique to identify latent subgroups that may vary in their levels of mediation. This investigation addresses this gap by simulating varying conditions of sample size, class separation, and mixing proportions for one-class, two-class, and three-class models. Information criteria (viz., AIC, BIC, aBIC) and likelihood ratio tests (viz., LMR, VLMR, and BLRT) will be evaluated for model selection accuracy. Preliminary results suggest that low sample sizes may reduce the accuracy of class enumeration. The finalized investigation aims to provide evidence that identifies conditions where the identification of latent classes is most accurate with mediation in regression mixture models.

Authors

• Natashia Bibriescas, University of Texas at Austin

• Tiffany Ann Whittaker, The University of Texas at Austin