In this article, we directly question the common practice in growth mixture model (GMM) applications that exclusively rely on the fitting model without covariates for GMM class enumeration. We provide theoretical and simulation evidence to demonstrate that exclusion of covariates from GMM class enumeration could be problematic in many cases. Based…

Using an approach to the analysis of missing data, this study investigated developmental trends in alcohol, marijuana, and cigarette use among 750 adolescents across 5 years using multiple-group latent growth modeling. Latent variable structural equation modeling and missing data approaches to studying developmental change are explored. (SLD)

Longitudinal data from a school-based alcohol-prevention trial involving 385 girls and 353 boys were used to show how a model using individual growth modeling and a covariance structure approach can investigate relationships among individual changes in two domains (positive and negative alcohol expectancies). Advantages of the approach are…

Several competing dynamic models were applied to adolescent substance abuse using a cohort of 3,580 participants in a longitudinal study beginning in grade five. Evidence was found for an intrinsic self-regulating mechanism in adolescent smoking and drinking. Support for a dampened linear oscillator model is discussed. (SLD)