The use of latent class and other mixture models is increasing rapidly in many social and behavioral sciences. This goals of this talk are to provide an overview of the use of the latent class model as both a psychometric measurement model and a model of change over time. Longitudinal reparameterizations of the latent class model have been developed in order to test specific sorts of hypotheses of change over time. The use of parameter restrictions to reduce model complexity and test hypotheses will be highlighted.