General theories of crime have frequently been used to explain a variety of offending and victimization experiences for a wide range of samples. However, feminist criminologists question whether the same causal mechanisms exert similar effects for males and females—a criticism that points to the need for sex-specific analyses. Toward that end, this dissertation examines variables derived from several different general theories of crime in three separate studies. Each of the studies uses split-sample analyses to investigate potential sex-based differences. The first study uses three-level meta-analytic methods to determine if predictor variables derived from general theories explain victimization for both adolescent males (n = 138,848) and adolescent females (n = 176,611). Additionally, it examines both within-dataset and between-dataset differences. The second study uses a sample of high school students in Arizona (n = 2,738 males, n = 2,932 females). It examines the role of parental social ties in explaining the overlap of adolescent dating violence (ADV) offending and victimization. The third study uses two waves of a longitudinal dataset of high-risk adolescents (n = 182 males, n = 203 females). It focuses on the relationship between negative emotions and delinquency, and the role of avoidant coping. In each of the studies, both gender-neutral and gender-specific explanations of offending and victimization were found. In the first study, while predictor variables derived from criminological theory explained victimization for both males and females, larger effect sizes were found for risky lifestyle variables. In the second study, an overlap between ADV offending and victimization was found for both males and females, and social ties explained some of the overlap. However, paternal attachment was only significant for females, and involvement was only significant for males. In the third study, avoidant coping was associated with an increase in substance abuse, and anger was associated with an increase in violent behavior for both males and females. Avoidant coping partially mediated the relationship between anger and substance use, but only for males. Implications for practice and future research are discussed.

Although young males are still the primary perpetrators of juvenile crime, girls are increasingly coming into contact with the criminal justice system. While girls may have different pathways to crime and risks for recidivism than boys, their risk to reoffend is typically assessed using a gender-neutral tool that is based on social learning theory: a theory originally developed and tested on males. With the appropriateness of using gender-neutral tools to assess female criminality coming into question, a number of researchers have searched for a resolution. To date, mixed findings on the predictive validity of risk assessment tools have not provided any definitive answers. To help assess the predictive validity of the Youth Level of Service Inventory, separate meta-analyses were conducted for male and female juvenile offenders using previous studies. The mean effect sizes were compared in order to determine whether the predictive validity is similar for both males and females. With the exception of violent recidivism, results indicate that the YLS/CMI works equally well for male and female offenders. The implications of these findings for theory, research, and correctional policy are discussed.