A Life-Course Approach to the Study of Victimization and Offending Behaviors

Abstract

Research generated from the life-course perspective has consistently documented a curvilinear relationship between age and offending, yet very little research has employed a life-course approach to investigate the relationship between age and victimization over time. We address this gap in the literature by using data from three waves of the National Longitudinal Study of Adolescent Health (Add Health). Specifically, based on Sampson and Laub's (1993) contention that the presence of adult social bonds decreases the likelihood of criminal activity in adulthood, we examine whether two adult social bonds—marriage and employment—are associated with desistance from delinquent involvement and whether these same bonds are associated with a reduction in the odds of victimization. The theoretical and policy implications of this research are discussed.

Keywords:

• victimization
• life-coures
• desistance
• delinquency
• informal social control

This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P01‐HD31921 from the National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Ronald R. Rindfuss and Barbara Entwisle deserve special acknowledgment for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 West Franklin Street, Chapel Hill, NC 27516 ([email protected]).

Notes

1. Others argue that the desistance process cannot be explained by informal social controls (Gottfredson & Hirschi, 1990), but rather can only be explained by changes in individual propensity. According to this view, criminal involvement is relatively unaffected by life-course transitions such as marriage and employment and any relationship between marriage and desistance is spurious. Although the Add Health data do now allow for a complete test of Gottfredson and Hirschi's (1990) assertions, we do include a measure of self-control to help take into account this competing explanation.

2. We should note that prior desistance researchers (e.g., Maume et al., 2005) examined whether there was a change in marital status between waves. We do not do this because at wave 2 the respondents were adolescents and thus very few reported being married. The same holds true for employment because only after graduating from high school does full-time employment likely occur. Thus, the need to examine changes in marital status or changes in employment status is minimal.

3. The selection bias term for employment for the delinquency sample included age, gender, race, the wave 1 delinquent peers, and the parental permissiveness scale. The selection bias term for employment for the victimization sample included age, gender, and the neighborhood problems scale. The selection bias term for marital status for the delinquency sample included age, gender, and race. The selection bias term for marital status for the victimization sample included age, gender, and race. Originally we had planned on applying propensity-score-matching (PSM) techniques to control for selection effects. However, PSM is largely based on the quality of the covariates included in the propensity scores (i.e., the selection bias term). Given that relatively few covariates were statistically significant, PSM is not the optimal choice and, as a result, was not employed in the current analyses.