For the purposes of this study, violent behavior was defined as "behaviors by individuals that intentionally threaten, attempt, or inflict physical harm on others." The study interviewed a sample of 717 male prison inmates drawn randomly from an intake cohort admitted to a State prison system over a 1-year period. Extensive background information was obtained for each respondent through the interviews and an examination of prison records. In order to explore the role that time-varying characteristics of individuals and their life circumstances played in violent behavior, researchers used a life events calendar to record month-by-month details about the respondents for the 3 years preceding their most recent arrest. Another component of the study examined violent events in the subjects' lives over the 3-year period, both as perpetrator and victim. Respondents were also asked to identify events in which there was a high risk of violence occurring, but in which the respondent neither attacked nor was attacked by anyone. Although most respondents were not incarcerated because of a violent offense, the majority of respondents reported some involvement in violence during the 3 years before their incarceration. A total of 502 respondents (70 percent) reported being involved in 1,664 assaults in which they either attacked someone or were themselves attacked. A total of 759 incidents with high risk for violence but in which it did not occur were reported by 572 respondents. Younger respondents were more likely to report assaults and a higher number of assaults, and early onset of criminal activity was a predictor of violent behavior. There was also a significant association between abuse as a child and the respondent's level of involvement in robberies. Further, when a respondent possessed a gun or other weapon, he was more likely to attack an opponent than when he did not have a weapon. The relationship between specific leisure activities and involvement in violence varied across different community contexts. Analyses have only begun to explore the effects of current life circumstances on involvement in violence. Respondents were more likely to report "costs too high" as a reason for avoidance when an incident involved strangers as opposed to acquaintances. Future research should focus on features that distinguish incidents with violent outcomes from those in which violence is avoided. 63 references and 25 tables