Stephanie Lanza

• Substance use behavior among adolescents and young adults
• Finite mixture modeling, including innovations in latent class analysis so that researchers can discover and understand hidden subgroups in their data
• Time-varying effect models to study substance use behavior developmentally and in real time/daily life
• Leveraging new technologies and methods to understand substance use behaviors in daily life based on intensive longitudinal assessment

Stephanie Lanza's career is devoted to advancing prevention research in behavioral health through the development and application of innovative quantitative methods. She focuses this work in two key areas. First, she works on extensions to latent class analysis so that researchers can discover hidden subgroups in their data and understand antecedents and consequences of their membership. Second, she focuses on translating time-varying effect models, which have been described in the statistical literature, for behavioral and health scientists and solve problems that arise in their application. These models enable scientists to examine regression coefficients as a continuous function of age or time, often using intensive longitudinal data. Her current research leverages new technologies and methods to understand substance use behaviors in daily life based on intensive longitudinal assessment.

Latent class analysis; time-varying effect modeling; longitudinal methods; etiology of health risk behaviors.