Police led pre-charge diversion models are an alternative response to traditional court procedures for young offenders under 18 years of age with limited or no prior involvement with the criminal justice system. Diversion occurs prior to charging and involves the police imposing sanctions such as a caution, restorative caution, final warning and reprimand. Pre-charge diversion models typically occur in the form of: i) diversion only, ii) diversion with referral to services (internal or external to criminal justice system), or iii) diversion with police-led restorative justice. These models start before the imposition of charges to avoid labelling young people as offenders, avoid exposing them to the formal criminal justice system, or by providing a mechanism of forgiveness and non-stigmatisation for reintegration purposes.This narrative is based on one meta-analytic review covering 19 studies. Eleven of the primary studies in the meta-analysis were based on evidence from the US, four from Australia, two from Canada, and two from the UK.
Overall, the evidence suggests that police led pre-charge diversion models for young offenders has reduced reoffending.
The meta-analysis found that police led pre-charge diversion models for young offenders led to a modest but statistically significant 6% decrease in reoffending compared to those who received standard formal sentencing. Assuming a 50% reoffending rate for those who received traditional processing, the average reoffending rate for those receiving pre-charge diversion was 44%.
Across the 19 studies, 31 individual comparisons of diversion versus traditional processing were identified as some studies included two or more diversion conditions compared to a single control condition. Eight found that police led pre-charge diversion led to a statistically significant decrease in reoffending, none reported an increase in reoffending and 23 found no evidence of effect compared to traditional processing. When the results are combined, the increased sample size provides a more precise estimate and overall statistically significant result.
After analysing the effect sizes by the quality of study methodology, the results from both the higher quality study designs (random assignment studies), and those of lower quality (quasi-experimental studies) show a decrease in reoffending for those who received pre-charge diversion compared to traditional processing. In both cases the effect is not statistically significant, but this is unsurprising given the small number of studies and the relatively modest overall result. Again, when the results of the two types of study design are combined, a result with greater precision is produced.
The Review was sufficiently systematic that many forms of bias that could influence the conclusions can be ruled out.
The evidence is taken from one meta-analytic review covering 19 studies, which demonstrated a high quality design in terms of having a transparent and well-designed search strategy, featuring a valid statistical analysis, sufficiently assessing publication bias, the risk of bias in analysis and considering the way outcomes are measured and combined. However, the Review did not sufficiently consider the influence of statistical outliers and did not appear to use an appropriate weighting scheme to account for less rigorous study designs.
Police led pre-charge diversion models for young offenders are assumed to reduce reoffending by:
However, these assumptions were not empirically tested. Future research should consider testing these potential change mechanisms.
The Review noted that the effect of the intervention might function according to a number of contextual conditions. These include:
The Review did not include detailed information relating to implementation of diversion models in the included studies, however it noted a number of elements specific to particular models:
The Review did not mention the costs or benefits of diversion models, and no formal economic analysis was provided.
There is evidence that police-led pre-charge diversion models for low-risk young offenders can modestly reduce crime overall. Diversion models are expected to prevent young offenders from being exposed to deviant peers in the criminal system and from being labelled as ‘delinquents’, which can increase the likelihood of reoffending. The Review concluded that there were no statistically significant differences in terms of reoffending between different diversion models or countries of study. Further research is needed to examine the implementation and economic issues surrounding police led pre-charge diversion for young offenders.
Review: Wilson, D.B., Brennan, I. and Olaghere, A. (2018), Police‐initiated diversion for youth to prevent future delinquent behavior: a systematic review. Campbell Systematic Reviews, 14: 1-88. doi:10.4073/csr.2018.5