Monte Carlo Simulation of Automated Growth Mixture Model Fitting and Heterogeneity Deductions

• In Event: AERA Virtual Poster Session 7
  In Poster Session: Poster Session

Abstract

This study investigated the performance of the two estimation methods (i.e., ML/EM and SEM Forest) in estimating the correct class enumeration when fitting one-class and three-class GMM under various experimental conditions by conducting two Monte Carlo studies. The findings of Study 1 and Study 2 revealed a remarkable divergence in the two estimators’ performances under the normal and nonnormal cells. Across the two studies, several unanticipated trends related to the influences of design factors on the two estimators’ performances were noted under nonnormal cells, precisely trends pertain to the sample size and class separation. This paper provided applied researchers with detailed descriptions of the performance of the two methods, building a bridge between confirmatory analyses and exploratory data mining algorithms

Author

• Amal Alhadabi, Kent State University - Kent