Assessing the Risk of Gang Members
Vol. 17 No. 3
Do gang members vary in their risk to commit crime?
Youth gangs have long been a problem in the United States and they have also become a major concern in Canada. The problem is especially salient in some of Canada's large urban centres where the media has given gangs prominent attention. Understandably, the increasing prevalence of gangs and media attention have fuelled the public's fear of gangs. For example, a poll conducted in the city of Montreal found 64% of respondents reporting that street gangs were a major problem in their city.
Various anti-gang initiatives have arisen over the years. They have ranged from specialized police squads to crime prevention efforts. The augmented efforts of police, prosecutors and the courts to deal swiftly and decisively with gang members have had two important consequences. First, there has been an influx of gang members into prisons that have created management and security concerns. Second, and this is common to all facets of the criminal justice system (i.e., prisons, courts, corrections), there is the need to differentiate gang members according to their risk to re-offend. Not all gang members may pose the same risk but it is unclear as to how best to differentiate gang members according to risk.
Drawing upon correctional data from one Canadian province, 86 male gang members were compared to 86 non-gang male offenders matched on age, city of residence, and status (custodial or community supervision). The gang members were identified using the criteria set out by the provincial police agency. All offenders were assessed using an evidence based actuarial risk/need instrument (see Research Summary, Vol. 13, No. 2, 2008). They were then followed up after their initial assessment for an average of 1,024 days and new convictions were recorded.
In terms of criminal histories, gang members were more likely than non-gang members to have committed crimes against the person (72% of gang members had been convicted of such a crime compared to 54% of non-gang members). Gang members were also more likely to be involved in prostitution offences although the incidence of prostitution offences was relatively low (8.1%). Interestingly, gang members did not have a history of more serious crimes but they did engage in a greater variety of crimes compared to non-gang offenders.
When overall risk/need was examined, almost 75% of gang members fell into the high or very high risk/need categories as measured by the assessment instrument (for non-gang members, approximately 63% scored in the high to very high range).
Furthermore, the specific risk/need factors that were relevant for non-gang members were also relevant for gang members with gang members scoring higher on factors such as criminal associations and a history of perpetration.
Finally, when the risk/need factors were summed to form a general risk score, this score predicted recidivism, whether any new conviction or a violent conviction, equally well with gang members as with non-gang members.
- Gang members vary in their risk to re-offend. Therefore, it is important that policy and practice avoid adopting a "one-size-fits-all" approach for addressing youth gangs.
- Risk/need assessments developed on the general offender population appear to apply equally well to gang members. Such instruments can be used to differentiate the risk levels of gang members and tailor interventions that are proportional to offender risk.
- The assessment of the criminogenic needs afforded by actuarial risk/need instruments can guide the selection of targets for treatment programs. Addressing the criminogenic needs of gang members through offender treatment programming can lead to reduced offender recidivism and improved public safety.
- Guay, J.-P. (2012). Predicting Recidivism with Street Gang Members (Corrections User Report 2012-02). Ottawa: Public Safety Canada.
For further information
James Bonta, Ph.D.
Public Safety Canada
340 Laurier Avenue West
Ottawa, Ontario K1A 0P8
Phone: 613-991-2831 Fax: 613-990-8295
- Date modified: