2011 data now included

Regular readers will know all about the Justice Data Lab, the Ministry of Justice pilot initiative which provides organisations working with offenders with the opportunity to evidence how effective their work is at reducing reoffending. To use the service, organisations simply provide details of the offenders who they have worked with and information about the services they have provided. The Lab then supplies aggregate one-year proven re-offending rates for that group, and, most importantly,  that of a matched control group of similar offenders.

The Justice Data Lab also now has access to reoffending data for individuals sentenced to community orders or released from prison in 2011 and so covers the full decade from 2002.  Organisations are invited to make new submissions if they would like the most up-to-date reoffending information on their service users.

The MoJ has committed to publish findings from the JDL on a monthly basis with reports issued on the second Thursday. Last Thursday’s edition (13 February 2014) included nine new reports, making a total of 55 so far. However, these nine reports relate to just four separate programmes with various sub-groups of Home Group’s offender support service considered separately. You can find details of all 55 reports on the dedicated MoJ web-page.

[divider]

The Findings

Unfortunately, this month’s report highlights some of the current difficulties with the JDL.

• In four out of nine cases, there was insufficient evidence to draw any conclusion about the impact of the programmes on reoffending.
• The JDL often finds it impossible to match the group of offenders of a service with an appropriate control group. In this case, the group of offenders that Home Group worked with could not be matched in terms of accommodation. Since the main purpose of the service was helping offenders to find accommodation this rendered the exercise pretty pointless and explained why in two of the sub-groups, offenders Home Group worked with were MORE likely to reoffend than those who didn’t receive a service.
• When positive results are found, cohorts are so small that the range of reduction in reoffending is too wide. This month Adelaide House women’s hostel reduced reoffending by between 1 and 30 percentage points and Warwickshire Youth Justice Service reduced reoffending by between 2 and 24 percentage points.

[divider]

Conclusion

Dspite the disappointing results to date, I remain a fan of the Justice Data Lab on the simple grounds that it is an important principle to make publicly available the effectiveness (or otherwise) of publicly funded reoffending programmes.

It is to be hoped that the ability to use 2011 data will enable organisations to submit much larger groups of offenders for the JDL to work on and for more meaningful results to emerge.

However, it does seem that the Data Lab is of little value for organisations working with drug or alcohol using or homeless offenders because, in these cases, the JDL cannot find a properly matched cohort.