ABSTRACT
According to general strain theory (GST), individuals engage in deviant behavior to reduce feelings of negative affect that occur because of strain. In this paper, we examine the relationship between a unique type of strain – sexual violence victimization – and several deviant outcomes. Drawing on GST, we hypothesize that sexual violence victimization will increase the odds of substance abuse, alcohol use, and violent/aggressive behavior and that feelings of anger, anxiety, and depression will mediate the relationship between these factors. To test these relationships, we use data from the Kaplan Longitudinal and Multigenerational Study. Results indicate that sexual violence victimization increases all three types of deviant behavior. Negative affect, however, does not mediate the relationship. We discuss potential explanations and suggestions for future research.
Acknowledgments
The authors gratefully acknowledge the Howard B. Kaplan Laboratory for Social Science Research in the Department of Sociology at Texas A&M University, College Station, Texas for providing access to KLAMS data.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. This is slightly more blacks and whites and fewer Hispanics than one would expect based on Houston’s demographic make-up between 1994-2002.
2. Within the KLAMS data set, questions about affect are contemporaneous rather than retrospective. Therefore, to reduce the number of confounds between the experience of strain and negative affect, experiences of sexual violence needed to be somewhat recent. Restricting the analysis to incidents that took place in the last year, however, would have reduced the sample size to just 102 respondents.
3. The Kuder-Richardson Reliability Coefficient is the equivalent of a Cronbach’s Alpha for dichotomous variables. The Polychoric Principal Component Analysis uses polychoric correlation coefficients, allowing the use of dichotomous items in this analysis.
4. Because the interactions between sexual violence and perceived parental rejection, low school commitment, and association with deviant peers were not statistically significant, we chose not to report the results from the models that included these interaction terms.