Recidivism in subgroups of serious juvenile offenders: Different profiles, different risks?

ABSTRACT Background Research has shown that the treatment of juvenile offenders is most effective when it takes into account the possible risk factors for re‐offending. It may be asked whether juvenile offenders can be treated as one homogeneous group, or, if they are divisible into subgroups, whether different risk factors are predictive of recidivism. Aims and hypotheses Our aims were to find out whether serious juvenile offenders may be subdivided into clearly defined subgroups and whether such subgroups might differ in terms of the risk factors that predict recidivism. Methods In a sample of 1111 serious juvenile offenders, latent class analysis was used to identify subgroups. For each juvenile offender, 70 risk factors were registered. Severity of recidivism was measured on a 12‐point scale. Analysis was then conducted to identify the risk factors that best predicted the different patterns of recidivism. Results Four distinct subgroups of juvenile offenders were identified: serious violent offenders, violent property offenders, property offenders, and sex offenders. Violent property offenders were the most serious recidivists and had the highest number of risk factors. Serious violent offenders and property offenders were characterised by overt and covert behaviour, respectively. Sex offenders differed from the other three groups in the rarity of their recidivism and in the risk factors that are present. For each of these four subgroups, a different set of risk factors was found to predict severity of recidivism. Conclusions Differences in recidivism rates occurred in spite of the fact that most of these youngsters had been in the standard treatment programme offered to serious juvenile offenders in the Netherlands. This was not a treatment outcome study, but the indication that two of the groups identified in our study appeared to be worse after going through this programme, whereas the other two did quite well in terms of recidivism lends weight to our idea that such classification of juvenile offenders may lead to more targeted treatment programmes that would better serve both the general public and the youths concerned. Copyright © 2011 John Wiley & Sons, Ltd.