Leveraging the pushes and pulls of gang disengagement to improve gang intervention: findings from three multi-site studies and a review of relevant gang programs

Abstract

The purpose of this article is to situate knowledge on the pushes and pulls of disengagement from gangs within the inventory of gang intervention programs. Drawing on developmental and life course criminological theory and three major, multi-site studies on gang disengagement, we examined the self-reported pushes and pulls that led gang members to reduce their gang embeddedness and move toward disengagement more effectively. We found that (1) multiple rather than single factors for leaving gangs were most common, (2) push factors exceeded pull factors in prevalence and frequency, and (3) motivations for disengagement may be age-graded, and appear to increase in complexity with age. We complemented the multi-site findings by examining prominent programmatic efforts to reduce or prevent gang involvement that have the most explicit theory of change related to the pushes and pulls in gang disengagement: focused deterrence, hospital-based interventions, jobs programs, and relationship-based interventions that have street-outreach, therapeutic, family, and fatherhood areas of focus. Programs that address individuals disillusioned with their gang and simultaneously offer sustained opportunities to develop and engage in prosocial networks are most likely to enjoy success. The success of such interventions, however, may be dependent on developmental stages in the life course.

Keywords:

• Disengagement
• desistance
• gang intervention
• motivations for gang exit

Acknowledgments

The authors would like to thank Dena Carson for her assistance in providing the relevant statistics for the G.R.E.A.T. II dataset. Data collection for the Connect Survey was supported by Award No. 2011-JV-FX-0105, from the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice. Funding from Google Ideas supported the data collection for the Google Ideas study. The content of this paper, however, is solely the responsibility of the authors and does not necessarily represent the official views of the U.S. Department of Justice or Google.

Notes

1. The Eurogang definition of gang membership has been widely endorsed and to be identified as a ‘Eurogang member’ a respondent must indicate that they have a group of friends, and that this group (1) is between ages 12 and 25 years, (2) spends time together in public places, (3) has been in existence for at least three months, (4) defines illegal activities as socially permissible, and (5) engages in illegal activities together.

2. We thank Dena Carson and her colleagues for providing us with supplemental analyses to populate the summary findings of Table 1. It should be noted that they examined whether motivations for leaving a gang differed by operational definitions of former gang membership, and found that although there were some differences, the overall patterns of findings with regard to frequency of responses were similar. We only draw on ‘operational definition 3’ because it most closely approaches the definitions used in the other data sets.

3. Although applicable to interventions with gang members, we do not include a review of comprehensive models that are designed to simultaneously address multiple ‘levels of practice’ (e.g., prevention, intervention, reentry, and suppression). Comprehensive interventions are highly context-specific, adapted differently across neighborhoods and cities. For instance, models such as Little Village/the Comprehensive Anti-Gang Model allow for variations in focus and balance across the prongs, specifically depending on context. Although when implemented with fidelity, evaluations of the Comprehensive Anti-Gang Model have shown positive impacts (Howell 2015; Spergel, Wa, and Sosa 2006), the multiple prongs make it difficult for us to associate specific mechanisms that could be leveraged for gang exit relative to motivations for leaving. However, undoubtedly, the Comprehensive Anti-Gang Model has components (e.g., street outreach) that could be highlighted relative to our findings.