Exploring intergenerational continuity in gang membership

Abstract

Little is known regarding intergenerational continuity in gang membership. Qualitative literature is suggestive of intergenerational parallelism yet no known research examines the causal mechanisms associated with this cycle, if it even exists. Prospective, longitudinal data from the Rochester Youth Development Study (RYDS) and the Rochester Intergenerational Study (RIGS) assess intergenerational continuity in gang membership among 371 parent–child dyads in a series of logistic regressions accounting for moderating influences of parent sex, child sex, parent–child sex combinations, and level of contact. Path analyses reported herein explore whether parenting behaviors mediate the relationship between parent and child gang membership among fathers and mothers, respectively. Three key findings emerge. First, intergenerational continuity in gang membership exists between mothers and daughters and, conditional on contact, between fathers and sons. Second, maltreatment mediates some of this relationship among father–son dyads. Third, no pathways to daughter gang membership were identified among mothers. In sum, this study provides evidence of intergenerational continuity in gang membership and further highlights the importance of parent sex, child sex, and level of contact in intergenerational research. Future research should further explore the causal pathways between parent and child gang membership.

Keywords:

• Gangs
• intergenerational continuity
• maltreatment
• gender

Notes

1. This is very similar to the arguments of Giordano (2010) who refers to direct transmission (i.e., modeling behaviors and conveying attitudes) and indirect transmission (i.e., providing poor environmental conditions and ineffective coping strategies) of antisocial behavior.

2. We rightfully acknowledge that we do not examine the role of IG transmission among adult onset gang members. In all likelihood, the causes of gang membership among those who join for the first time at age 18 or older are likely different from those who join a gang for the first time as a juvenile.

3. The age of first self-reported gang membership is also used to ensure temporal ordering between our independent variable, mediators, moderators, and G3 gang membership.

4. Additional information was collected indicating the age of G3 at maltreatment. This was used to ensure temporal ordering so that G2 maltreatment of G3 was temporally prior to G3 joining a gang.

5. There is an element of right-censoring in the measurement of maltreatment perpetration and victimization. If a G2–G3 dyad moved out of New York state, we are unable to assess maltreatment perpetration and victimization. However, the implication of this limitation is that we falsely identify G2s as not perpetrating an act of maltreatment against G3. This would mean that our estimates are biased downward and we are presenting a conservative estimate of the relationship between G2 gang membership, G2 perpetration of maltreatment of G3, and G3 gang membership.

6. Prior intergenerational research looking at time-stable information regarding parent–child dyads generated from prospective, longitudinal design originating from one birth cohort accounts for the parent age at birth of the child in two ways: (1) as a control variable or (2) as a mediator of the relationship between parental behavior and child behavior.  Therefore, our analyses were run both ways to determine the sensitivity to methodological decisions. In both analyses, the results on our variables of interest in terms of magnitude and significance are the same.  Moreover, age at birth of the child was not a significant mediator of the relationship between parent gang membership and child gang membership, but it is significant in some of the individual paths estimated.  Thus, we present the models controlling for age at birth only.