The research sought to determine whether the closely related lifestyle and routine activities perspectives on personal victimization could provide the basic elements of a comprehensive, multilevel framework that balances human agency and lifestyle choices against social determination and balances social control against social attachment. The data were collected in January 1995 by means of anonymous questionnaires distributed in school classes. Responses came from 1,879 females and 1,931 males, who represented 87 percent of the Icelanders in this age group. Results were consistent with findings in the United States and the United Kingdom in that violent offenders had the highest risk of violent victimization; the risk increased with higher rates of violent offending. The use of violence was related to higher rates of alcohol use and to higher rates of theft and vandalism, supporting the argument of Sampson and Lauritsen that violent lifestyles are embedded within more general delinquent lifestyles. Females were less violent and were less likely to be victimized by nonsexual violence. Results were also consistent with Jensen and Brownfield's argument that the observed gender differences in victimization can be traced to gender differences in exposure, delinquency, and guardianship. Findings underscore the importance of a comprehensive synthesis of delinquency and victimization theories, which could specify the complex relationships between human agency and structural constraints. Tables, footnotes, and 37 references