Latent Transition Analysis With Random Intercepts (RI-LTA)

This article demonstrates that the regular LTA model is unnecessarily restrictive and that an alternative model is readily available that typically fits the data much better, leads to better estimates of the transition probabilities, and extracts new information from the data. By allowing random intercept variation in the model, between-subject variation is separated from the within-subject latent class transitions over time allowing a clearer interpretation of the data. Analysis of two examples from the literature demonstrates the advantages of random intercept LTA. Model variations include Mover-Stayer analysis, measurement invariance analysis, and analysis with covariates. Translational Abstract Modeling with latent classes over time is a common approach in psychology when studying the development of for example mental states of happiness or depression over time. Latent transition analysis is a well-known approach for this purpose. A better statistical approach is presented here which represents the data better and more correctly assesses change and stability over time. Interpretations of psychological change processes are changed by this new methodology. Earlier LTA findings need to be revisited.