The purpose of this study was to examine the utility of social control theory's "attachment" and "involvement" bonds as protective factors to examine gender differences in reports of delinquency and academic achievement in a sample of 7th to 12th graders. Attachment bonds were operationalized as attachment to parents, to nonparent adults, and to peers; involvement bonds were operationalized as time spent in various school- and non-school-based activities. It was expected that, with respect to delinquency and academic achievement, attachment bond variables would be more protective for females than the involvement bond variables, whereas the involvement bond variables would be more protective for males than the attachment bond variables. The findings suggest that although several of the involvement bond variables of social control theory are predictive of both delinquency and academic achievement for both genders, only the attachment bond variables provide such an overall protective function for females. Implications for theory and practice are discussed.