![Sample our Law journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5942/TOC-Subject-Sample-2021-Law.jpg"})
Abstract
Previous examinations of co‐offending have identified a subset of high‐rate offenders who commit crimes with a large number of co‐offenders, most of whom are younger and less criminally experienced. These so‐called “recruiters” are of particular interest to researchers and practitioners, because of their potential role in facilitating offending onset and recidivism among their co‐offenders. In this paper, data on 61,646 individuals detected by a large UK police force are used to identify offenders who fitted the recruiter profile, and to compare their individual and offending characteristics with those of non‐recruiters. In total, 86 recruiters were identified. In multivariate analyses, recruiters were found to be older than non‐recruiters and were typically involved in property crimes. In addition, they tended to offend in criminal groups that were more heterogeneous and stable than non‐recruiters. These findings suggest that a small but identifiable group of recruiters can be detected using official data and that these individuals may be important targets for police attention and court treatment.
Acknowledgments
A previous version of this paper was presented at the 59th annual meeting of the American Society of Criminology in November 2007. The authors are grateful to the police for providing the data analyzed herein, and to Peter Carrington, Kevin Conway, the IoC writing group, and three anonymous referees for providing valuable comments on earlier drafts of this paper. This research was supported by the Gates Cambridge Trust, the UK Economic and Social Research Council (PTA‐026‐27‐1744), and the Danish Research Council for the Humanities.
Notes
1. In this paper, we adopt a traditional definition of co‐offending as the act of committing crime alongside one or more accomplices (Reiss, Citation1988).
2. This dataset was made available to the authors as part of a larger study of co‐offending patterns in UK police data (see van Mastrigt & Farrington, Citation2009).
3. Ten years is the minimum age of criminal responsibility in the UK. Offenders under this age were excluded from analysis.
4. Carrington’s (Citation2002, Citation2009) Canadian UCR2 data is another excellent source of official data on co‐offending.
5. Due to space constraints, only the multivariate tests are reported for the secondary analyses presented in Appendix 1.
6. Some of these variables were excluded in the later logistic regression analysis due to problems with multicollinearity.
7. In the secondary analyses presented in Appendix 1, a different conceptualization of prolific offending was used in order to account for the relative frequency of different offence types in the data. Because the opportunity structures and, in turn, the relative frequencies of different crime types can vary, some critics might argue that committing 10 violent offenses is not equivalent to committing 10 property offenses or 10 drug offenses. According to this logic, a general requirement of 10 offences may be too restrictive, as offenders with less than 10 offenses may still be considered prolific relative to their peers depending on the types of offenses they commit (the authors are grateful to an anonymous reviewer for raising this issue). In order to address this possibility, an offender was considered prolific in the secondary analyses if s/he fell into the extreme upper tail of the frequency distribution for one or more general offending categories (violence, property, drugs), n = 2,329. For further details regarding the secondary selection procedure, see Appendix 1.
8. In addition to committing significantly more offences (15.2 vs. 1.65, on average) prolific offenders were younger than their non‐prolific counterparts (23.5 years vs. 26.7, p < .01), and were more likely to be male (88.2% vs. 78.2%), χ 2 = 78.9, p < .001, OR = 2.09 (95%CI = 1.77–2.47).
9. On average, prolific offenders were linked to 2.71 unique co‐offenders each.
10. n = 356 in the secondary analysis (see Appendix 1).
11. n = 105 in the secondary analysis (see Appendix 1).
12. Similar findings emerged in the secondary analysis presented in Appendix 1; however, age fell just short of statistical significance in that model.
13. As above, age failed to reach significance as an independent predictor in the secondary model. However, number of offences and number of 6‐month offending periods maintained significance in distinguishing between recruiters and prolific offenders. This highlights the importance of considering how alternative definitions of recruitment may impact on the relationships observed.
14. The same results emerged from the secondary analysis presented in Appendix 1.
15. This estimate is restricted to males, as only a small proportion of recruiters are female.
16. Readers should note that these estimates are based on the assumption that the targeted individuals are “pure instigators,” that is, recruiters rather than joiners for all of their offences. The extent to which this might be the case in reality remains unclear (particularly in light of our finding that many recruiters’ accomplices were themselves relatively active offenders). These figures should thus be interpreted as estimates of the maximum number of crimes saved. Until more comprehensive and precise information is available on the criminal careers of recruiters and their accomplices and, in particular, the transiency of recruiter/joiner statuses, any estimates of crimes saved will be necessarily rough.
