Based on predictors that were found to be associated with male offending and female offending, some theories of delinquency were apparently better suited to explaining either male or female delinquent behavior. Also parts of one theory of delinquency may apply to one gender and other parts to the other gender. The analysis of gender-related predictors of delinquency is further complicated by differences in predictors for specific types of delinquent behavior and how these may differ by gender with reference to each type of delinquent behavior. Consequently, general theories of delinquency fail to address differences in risk factors by type of delinquent behavior as well as differences in predictors by gender for each type of delinquent behavior. In order to determine whether gender-specific theories are needed for specific types of delinquent behavior, tests of competing theories must be conducted in the same models using various outcome measures. The current study used data from the National Longitudinal Survey of Adolescent Health. The analysis used variables not only from traditional criminological theories (strain, social bond, and differential association and social learning theory) and the newer life-course approach, but also from the feminist perspective. The survey is a school-based, nationally representative multiwave panel study of adolescents in grades 7 through 12 in the United States. Measures address social and demographic characteristics, household structure, parenting and family dynamics, risky behaviors, decisionmaking, self-esteem, relationships, education and employment, and health. Two of the three waves of data were used (n=3,422). Wave I was conducted from September 1994 through December 1995, and Wave II were conducted between April 1996 and August 1996. 4 tables, 18 notes, and 95 references