The first chapter presents an introduction to the NHSDA methodology and the organization of the risk and protective factors included in this report. Risk factors are those individual and social-environment characteristics associated with an increased likelihood of substance use; and protective factors are those features of an individual and social environment that increase the likelihood that a juvenile will abstain from substance use. In examining the distributions of risk and protective factors for substance use, this analysis focuses on four domains based on categories developed by the Social Development Research Group of the School of Social Work at the University of Washington in Seattle. The community domain includes such factors as community disorganization and crime and the availability of illicit drugs. The family domain encompasses such factors as parental attitudes toward youth substance use and parental communication with youth about the dangers of substance use. The peer/individual domain includes such factors as sanctions against substance use at school and exposure to prevention messages in school. The second chapter examines the prevalence of the various risk and protective factors in the population as a whole and by demographic variables such as race/ethnicity, gender, and age. The third chapter examines the associations between these factors and past-year marijuana use. Chapter 4 considers the relative predictive power of risk and protective factors, both individual and by domain, in a multiple logistic regression analysis. It also develops the best model for "predicting" juvenile marijuana use based on a reduced set of risk and protective factors from all four domains. Further, it explores the extension of modeling to explain variation at higher levels of the hierarchical structure (i.e., families and communities), including a brief introduction to hierarchical modeling. Models of cigarette use and alcohol use are presented in selected tables. The fifth chapter discusses methods and issues associated with measuring the change in risk and protective factors over time, and it compares the distributions of risk and protective factors from 1997 to 1999, as well as the associations between these risk and protective factors and marijuana use for these 2 years. It also discusses how changes in the distribution of risk and protective factors and changes in the strength of association between the risk and protective factors and marijuana use between 1997 and 1999 may explain the decrease in juvenile marijuana prevalence rates during this period. Final conclusions of the study are presented in chapter 6. Among the four domains of risk and protective factors, the peer/individual domain explained the largest amount of variation, with the strongest predictors being participation in antisocial behavior, friends' marijuana use, low perceived risk of marijuana use, and positive attitude toward marijuana use. The finding that the peer/individual domain contained the strongest predictors of youth marijuana use was consistent across both the 1997 and 1999 surveys. Extensive tables and figures, 52 references, and appendices that focus on various aspects of methodology