Adult and Youth Reoffending in Northern Ireland (Cohort 2012 to 13) Published Today

The Department of Justice (DOJ) today published Research and Statistics Bulletins: 16/2015 ‘Northern Ireland Reoffending Methodology: Methodology and Glossary Part 2 (New Edition: August 2015) and 17/2015 ‘Adult and Youth Reoffending In Northern Ireland (2012/13 Cohort)’.

These are Official Statistics publications. Bulletin 16/2015 provides a discussion on how overall reoffending rates can be compared.   This is highly technical but necessary to provide clear understanding on how appropriate comparisons can be made.

In addition to this methodological discussion, a second paper has also been published providing proven reoffending rates both for adults and for youths aged 17 and under who have been given a non-custodial disposal at court, a diversionary disposal or who have been released from custody during 2012/13.  The main findings of each report are presented below.

Research and Statistical Bulletin 16/2015: Northern Ireland Reoffending Methodology: Methodology and Glossary Part 2 (New Edition: August 2015)

• The first of the two papers published examines the methodology for adjusting actual reoffending rates to aid comparison across different yearly cohorts. This is highly technical but necessary to provide a clear understanding on how appropriate comparisons can be made.  This is a new edition of the second methodology paper, originally published in December 2014, and was necessitated primarily due to a change in sentencing practice with the introduction of Penalty Notices for Disorder in June 2012.  After such changes it was important to review the existing model to ensure continued accuracy.  On further examination it was determined that a new model was needed and was created using the same methodology as before. This new edition therefore presents information based on this new model.
• Using logistic regression a suitable model was found for the overall cohort of adults and youths combined.  This model was also applicable for the adult cohort considered separately.
• It should be noted that the model devised does not take into consideration any changes in legislation, sentencing patterns, emerging offences, treatment practices or dynamic factors associated with reoffending. It is therefore important that the model be periodically reviewed to ensure that it is still fit for purpose.
• By comparing the actual with the predicted scores we can examine the change in reoffending beyond what we would have expected from those individuals within the cohort. For example, although the actual reoffending rates increased by approximately 2 percentage points between the 2010/11 and 2012/13 years, we might have expected an increase of approximately 4 percentage points based on the characteristics of those within the cohorts.  Overall this suggests that after controlling for offender characteristics reoffending actually decreased by approximately 2 percentage points between 2010/11 and 2012/13.
• At this point no accurate model of youth reoffending could be found.  As more information becomes available this will be reviewed.

Research and Statistical Bulletin 17/2015: Adult and Youth Reoffending in Northern Ireland (2012/13 Cohort)

• This bulletin provides information on the one year proven reoffending rate for offenders who received a non-custodial disposal at court, a diversionary disposal or were released from custody during 2012/13.  Information is presented in relation to the full cohort and also broken down by adults (those aged 18 and over) and youths (those aged 17 and under).
• Of the 29,427 people included in the 2012/13 cohort, 5,452 (18.5%) reoffended during the one year observational period.
• Of the 5,452 who reoffended, over two fifths (41.9%) committed their first reoffence within the first three months.
• In terms of offending history, 60.0% had committed previous offences ranging from one to 653 distinct offences. Reoffending rates increased with the number of previous offences.
• Overall, 12.2% of females and 20.1% of males had reoffended.
• Base reoffending rates should not be used to measure the comparative success of different disposal types in their own right. The reason for this is that different offender characteristics and histories and different offence types will typically themselves be related to the type of disposal previously given. Therefore offender profiles may differ substantially between the different disposal types. Subject to this necessary caveat: The one year proven reoffending rate for those released from custody was 46.8%. The one year proven reoffending rate for those who received a community disposal at court requiring supervision was 34.2%. The one year proven reoffending rate for those who received a community disposal at court not requiring supervision was 18.4%. The one year proven reoffending rate for those who received a diversionary disposal was 17.2%.
• The highest reoffending rates were found amongst those who had committed a baseline offence in the ‘Burglary’ category (38.2%), followed by ‘Robbery’ (37.5%).

Adult Cohort

• Of the 27,130 adult offenders included in the 2012/13 cohort, 4,760 (17.5%) committed a proven reoffence within a year following being released from custody, receiving a non-custodial disposal at court or receiving a diversionary disposal.
• Overall, 41.4% of the 4,760 who reoffended committed their first reoffence within the first three months of the observation period.
• In terms of offending history, 61.2% had committed previous offences, ranging from 1 to 653 distinct offences.  Reoffending rates largely increased with the number of previous offences.
• Overall, 11.4% of females and 19.0% of males had reoffended.
• Subject to the caveat about comparing base offending rates previously given
• The one year proven reoffending rate for adults released from custody was 45.8%.
• The one year proven reoffending rate for adults who received a community disposal at court requiring supervision was 30.7%.
• The one year proven reoffending rate for adults who received a community disposal at court not requiring supervision was 18.0%.
• The one year proven reoffending rate for adults who received a diversionary disposal was 14.3%.
• The highest reoffending rates were found amongst those who had committed a baseline offence in the ‘Burglary’ category (39.5%), followed by ‘Robbery’ (35.0%).

Youth Cohort

• Of the 2,297 young offenders included in the 2012/13 youth cohort, 692 (30.1%) committed a proven reoffence within a year following being released from custody, receiving a non-custodial disposal at court or receiving a diversionary disposal.
• Almost half (45.2%) of the 692 who reoffended committed their first reoffence within the first three months of the observation period.
• In terms of offending history, 46.3% had committed previous offences ranging from one to 92 distinct offences. Reoffending rates increased with the number of previous offences.
• Overall, 21.2% of females and 32.4% of males had reoffended.
• Subject to the caveat about comparing base offending rates previously given:
• Of the 40 youths released from custody, 34 committed a proven reoffence.
• The one year proven reoffending rate for youths who received a community disposal at court requiring supervision was 55.5%.
• The one year proven reoffending rate for youths who received a community disposal at court not requiring supervision was 51.0%.
• The one year proven reoffending rate for youths who received a diversionary disposal was 27.0%
• Although the numbers are small, seven of the 12 youths who committed a baseline offence in the ‘Robbery’ category reoffended, followed by ‘Possession of Weapons’ (39.2%) and ‘Criminal Damage’ (37.6%).

Notes to editors: 

1. In 2013, the Department of Justice’s Analytical Services Group embarked on a project to revise the methodology used to calculate recidivism rates within Northern Ireland, bringing it more in line with established methodology in England and Wales. For a more detailed methodology refer to ‘Northern Ireland Reoffending Methodology: Methodology and Glossary Part 1’.
2. Any study of reoffending rates is prone to misunderstanding and misrepresentation if sufficient care is not taken to observe the caveats around each figure. For example, for both adult and youth cases here, the reoffending rates are highest for those released from custody and lowest for those given a diversionary disposal. Inevitably what needs to be taken into account in the interpretation of these figures is, most obviously, (a) the seriousness of the offence which led to the disposal in the first place and (b) the previous criminal history of the individual as a factor in the original disposal, together with a range of other criminogenic, demographic and, indeed, administrative/procedural issues. What these figures do not mean, of course, is that diversionary disposals are, irrespective of other factors, necessarily a more efficient deterrent to reoffending.
3. A range of statistical controls and procedures are therefore required before comparisons can meaningfully be made across different reoffending rates. This second methodology paper therefore provides a discussion on how overall reoffending rates can be compared.  
4. This paper is a new edition of the second methodology paper, originally published in December 2014, and, like the first edition, provides a discussion of the difficulties surrounding the comparison of yearly reoffending rates and guidance on how this can be achieved.  
5. Examination of the overall cohorts revealed that there were substantially fewer community sentences that do not require supervision and diversions in the 2012/13 cohort.  A major factor impacting this may well be the result of the introduction of Penalty Notices for Disorder (PND) in June 2012.  After such changes in sentencing practices it is important to review the established model to ensure accuracy.  On further examination it was determined that a new model was needed and was created using the same methodology as before. This new edition therefore presents information based on this new model.
6. Also provided are a further year’s proven reoffending rates for both adults and youths.
7. Official Statistics are produced in accordance with the Code of Practice for Official Statistics.  They undergo regular quality assurance reviews to ensure that they meet customer needs and are produced free from any political interference.  They are also subject to restrictions in terms of pre-release access.
8. The statistics bulletin is available from the publication section alternatively contact;

   Analytical Services Group
   Laganside House
   Oxford Street
   Belfast
   BT1 3LA.

Email: statistics.research@dojni.x.gsi.gov.uk.

9.  Press queries about this publication should be directed to the DOJ Press Office on telephone number 028 9052 6444.

Share this page