Two-step growth mixture model to examine heterogeneity in nonlinear trajectories

Empirical researchers are usually interested in investigating the impacts that baseline covariates have when uncovering sample heterogeneity and separating samples into more homogeneous groups. However, a considerable number of studies structural equation modeling (SEM) framework start with vague hypotheses terms possible causes. It suggests (1) determination specification proper model is not straightforward, (2) exploration process may be computationally intensive given SEM complicated pool candidate huge psychological educational domain where widely employed. Following Bakk Kuha (2017), this article presents two-step growth mixture (GMM) examines relationship between latent classes nonlinear trajectories characteristics. Our simulation demonstrate proposed capable clustering change patterns, estimating parameters interest unbiasedly, precisely, as well exhibiting appropriate confidence interval coverage. Considering highly correlated, study also proposes implementing exploratory factor analysis (EFA) to reduce dimension covariate space. We illustrate how use hybrid method, GMM EFA, efficiently explore longitudinal mathematics achievement data.