How to model intraindividual change in cohort data using Mplus’ multi-group approach

The present article aims to show how to model longitudinal change in cohort sequential data applying latent true change models using Mplus’ multi-group approach. The underlying modeling ideas are described and explained in this article. As an example, change in internalizing problem behaviors between the age of 8 and 13 years is modeled and predicted by gender. The example data stems from a large German cohort sequential study and comprises 806 children in three birth cohorts (2001, 2000, 1999). Finally, the advantages and disadvantages of this modelling approach are discussed. Annotated syntax is provided online for the models.