This overview describes a distinct semi-parametric, group-based approach for modeling developmental trajectories. The method is based on "finite mixture modeling." This method was used with a Montreal-based (Canada) dataset obtained from tracking 1,037 White males of French ancestry. The subjects were selected in 1984 from kindergarten classes in low socioeconomic neighborhoods. Following an assessment at age 6, the boys and other informants were interviewed annually from the ages of 10 to 17. Assessments were done on a wide range of factors. Among these were self-reports on whether or not the boy had been involved with a delinquent gang in the past year. Application of the group-based method to this gang-involvement data identified three clearly distinct groups. The trajectory for each group was described by the probability of gang membership at each age. After describing the research methodology, this overview explains rationales for using a group-based statistical model to examine both normal and pathological development. The author advises that much work remains to be done in improving group-based trajectory modeling, with perhaps the most important being to provide the capacity for taking into account the possibility that turning-point events may be both a cause and effect of the behavior being studied. Another important issue is the development of guidelines for making well-informed judgments about the optimal number of groups to be included in a research project. 4 figures and 37 references