Individual papers focus on new statistical methods for drug prevention research, the role of missing data in analyses of drug prevention, latent-class analysis of drug abuse patterns, and the use of latent transition analysis in addressing questions in prevention research. Some papers use examples from marijuana research to discuss the incorporation of trend data in causal interpretations of individual-level correlations among variables, the potential applications of multilevel models for hierarchically nested data, ways of improving analyses, and the use of survival analysis in prevention research. Other papers focus on time series models of individual drug abusers, the use and misuse of repeated measures designs, meta-analytical issues, and dynamic systems-modeling as a means to estimate the effects of community-based prevention. Tables, figures, and chapter reference lists