Data for the study were obtained from the first five waves of the Denver Youth Survey (DYS), a longitudinal household-based sample of children and adolescents who lived in high-risk neighborhoods in Denver in 1987. High-risk neighborhoods are defined as areas characterized by both social disorganization and high official crime rates. Factor and cluster analyses were used to identify socially disorganized areas of Denver. Then, using arrest data, those neighborhoods within the socially disorganized areas that were in the upper one-third of the crime distribution were selected for inclusion in the DYS study. The overall design of the research project was based on an accelerated longitudinal research design. Self-reported Total Victimization was a composite measure of six items asked of youth respondents. Measures of annual prevalence (ever during the past year) and annual frequency of victimization (the number of times the respondent was victimized during the past year, the basic measure of multiple victimization) were obtained from respondents in all of the first 5 years of the survey. Based on the annual prevalence and frequency data, cumulative prevalence and frequency of victimization were constructed. Over a 5-year period, the results showed that chronic, multiple intermittent victimization was the usual pattern among respondents in the sample. 1 figure, 6 tables, 3 notes, and 29 references