Epidemiological research shows some ethnic-group differences in the prevalence of substance use. This approach does not address the question of whether there are ethnic-group differences in the models that are needed to understand the development of substance use. For this question we need to understand the relations between psychological constructs and their trajectories over time. In this paper we describe some approaches to studying ethnic-group differences in the predictors of substance use that illustrate probing for mediators, multisample analyses of structural models, and an experimental trial of a preventive intervention. Our studies found some ethnic-group differences in the structure of constructs and the relations between variables, but many similarities. The challenge for researchers is using appropriate research methods for studying ethnicity, uncovering the basis for ethnic-group differences when they occur, knowing when statistical differences are meaningful, and acknowledging when developmental models are comparable.