Longitudinal research designs involve repeated measurement over time of one or more groups of subjects. The major advantage of such designs is the ability to study a phenomenon's natural history and course of development. Longitudinal research design types include trend studies, cohort studies, panel designs, and time-series designs. Choosing between different design types involves issues related to the sample versus population, prospective versus retrospective studies, data sources, and correlation versus causation. Examples of each longitudinal research design type can be found in criminology. Over the past two decades, however, longitudinal research has focused on criminal careers using primarily cohort and panel designs. Further, the choice of longitudinal versus cross-sectional research design in criminology has been debated since 1983. A key factor in this debate appears to be the continued emphasis by government funding sources, such as the National Institute of Justice and the Office of Juvenile Justice and Delinquency Prevention, on the need for longitudinal studies. The author concludes that longitudinal designs are best suited for causal development because they involve repeated measurements and facilitate the analysis of change patterns. 36 references