The introductory section discusses the rationale for the study of longitudinal methods, a type of study uniquely suited to identifying factors affecting the emergence or nonemergence of deviant behavior from the perinatal period. Information from such studies could be of particular importance for devising primary crime prevention measures. Such prevention focuses on the early identification of those in whom some interaction of biology, psychology, and social conditions produces a predisposition to later antisocial behavior. Following a discussion of the history of longitudinal research, an overview covers methodological and practical problems in research strategies, along with longitudinal methods in the study of development. The section on types of longitudinal research designs examines sequential strategies and the role of cohort effects in behavioral development, with attention to the adolescent personality. Structural equation models in longitudinal research and an application of causal modeling techniques to prospective longitudinal data bases are considered in the discussion of statistical approaches. A discussion of followup procedures considers minimizing subject attrition and usefulness of official records, while that on confidentiality and ethics describes the American Council on Education link file, steps in the research process that could insure right to privacy, informed consent, deception regarding research purpose, and more. A final chapter looks at specific questions in criminologic practice and theory and their suitability to longitudinal research. Chapter references are included. For separate documents in the presentation, see NCJ 83032-33.