An examination of parameter recovery in latent transition models with distal outcomes.

Latent transition analysis (LTA) is an increasingly popular research method used to categorize subsets of individuals within a population. The current study is purposed to investigate parameter recovery of a distal outcome effect in an LTA model. All models have a two-class solution with two time points. Design factors of interest include sample size, class prevalences at Time 1, the transition parameter, and the distal outcome effect size. ANOVA was used to seek out any practically significant effects (η2P ≥ 0.14) of these design factors on the raw bias (RB) of the distal outcome estimate. The results revealed no practically significant effects, meaning that the LTA model accurately estimated the distal outcome effect size under all specified conditions. Future research could expand upon this study by including different numbers of classes, indicators, time points, distal outcomes, and other auxiliary variables in addition to the distal outcome(s).