Latent Transition Analysis for Program Evaluation with Multivariate Longitudinal Outcomes

Evaluations of intervention programs, such as the PAX Good Behaviour Game (PAX) program often havemultiple outcome variables (e.g., emotional symptoms, conduct problems, hyperactivity/ inattention, peer relationship problems, and prosocial behaviour). These are often reported for multiple time points (e.g., preand post-intervention) where data are multilevel (e.g., students nested in schools). In this paper, we use latent transition analysis (LTA), a person-oriented statistical approach, to evaluate the PAX program with multilevel, longitudinal multivariate outcomes. Using data from the Manitoba PAX Study, we show how LTA helps explore the transition of multiple outcomes across multiple time points and how the intervention program affects this transition. The strengths and limitations of LTA are discussed.