Centenial Celebration

Transaction Search Form: please type in any of the fields below.

Date: November 22, 2024 Fri

Time: 11:37 am

Results for community-based corrections cognitive skills

1 results found

Author: Turner, Russell

Title: Thinking about Re-offending: Reductions in reconviction rates for offenders on the Enhanced Thinking Skills (ETS) programme.

Summary: The Enhanced Thinking Skills (ETS) programme has been operating in West Yorkshire for over 5 years and 2005/6 saw 882 offenders ordered to attend the programme. This is a sizeable investment in accredited programmes as a method to tackle general offending behaviour. National research on the effectiveness of programmes such as ETS is still in its infancy and at present gives mixed results. How well an area delivers its programmes may in part cause this. There has been much local research in West Yorkshire on programme implementation and delivery but none on the impact on reconviction. This report aimed to provide data on the impact upon reconviction during a two-year follow-up period on a sample of 298 offenders ordered to attend the programme mainly during 2002/3 as part of a Community Rehabilitation Order (CRO) or Community Punishment & Rehabilitation Order. Reconviction data was also obtained on a comparison group of 182 offenders who were either eligible to attend the programme or likely to have been eligible to attend, but were given a CRO without an ETS condition during the same period. Reconviction data from the Home Office could not be matched for 11% (n=51) of the cases: 9% (n=28) from the ETS sample and 13% (n=23) from the comparison group. The remaining samples – 264 ETS offenders and 157 comparison offenders – differed on several key aspects. Predicted reconviction rates, as given by the Offender Group Reconviction Scale (OGRS), were lower for the comparison sample at 50.5, compared to 63.6 for the ETS sample. The comparison sample was on average slightly older with an average age of 28.6 compared to 26.4, and proportionately had more women: 29% (n=46) compared to 13% (n=34) in the ETS sample. Lower predicted rates were also found for the comparison group when men and women were considered separately. The relationship between these factors - OGRS, age, and gender - and reconviction is complex but taken together strongly suggests that we would expect the comparison sample to have lower reconviction rates than the ETS sample, even if the ETS sample do better than expected using their predicted reconviction rate. Conversely, the ETS sample had a lower actual reconviction rate than the comparison sample: 60.6% were reconvicted within two years compared to 66.2% of the comparison sample, a difference of 5.6%. This difference was not sufficient to be statistically significant with this sample and may have arisen due to chance. The actual reconviction rate for the ETS sample was also 3% lower than their predicted rate, whilst the rate for the comparison group was 16% higher than their predicted rate. A ‘true’ reduction for the ETS group, from predicted to actual reconviction rate, is very likely to be between 0.6% and 5.4%, based on this sample. Clearly, this ranges between a negligible effect to one in line with national targets, i.e. the Government’s 5% reduction in reconviction. The 3% difference found for this West Yorkshire sample might be indicative of a ‘real’ difference in the behaviour of these offenders, compared to what was expected. However, these rates are based on all offenders sentenced to ETS. The reconviction rate halves to 33% for those offenders who complete the programme, compared to 73% for the combined non-completers. This difference is statistically significant and also stands when differences in predicted reconviction rates between the completers and non-completers are taken into account. Additionally, the completers’ actual reconviction rate is 25% points lower than their predicted rate, whereas the non-completers do show any improvement effect. This strongly suggests that the ETS programme in West Yorkshire helps reduce reconviction providing offenders complete the programme. What this research does not address however is the relationship between a ‘treated’ offender profile (see Friendship et al., 2003) and reductions in reconvictions, i.e. to what extent did ETS change offenders’ thinking, and their subsequent behaviour? A third of offenders sentenced to ETS had OGRS scores over 75. These offenders may have had additional conditions imposed to address their risk of re-offending, as opposed to simply being poor-targeting. When different risk-bands are considered separately, only the medium-high risk offenders reconvict less than expected. The medium-low offenders’ reconviction rate is 5.5% higher than that given by OGRS, although it is 9% lower than the comparison group. The medium-high risk offenders, on the other hand, have an actual rate some 8% lower than the predicted rate and 25% lower than the comparison group. Whilst this implies that ETS has a stronger effect on medium-high risk offenders, it reaffirms the need for correct targeting. Whilst reconviction rates for each age-band in the ETS sample are between 2% and 7% lower than the comparison sample, they are not lower than the predicted rates in all age-bands. Indeed, only the 18-20 age-group shows a reliable effect with the reconviction rate for this group being nearly 9% lower than predicted. The other three age-groups do not show any notable effect. This suggests that the ETS programme in West Yorkshire had a greater impact upon reconviction rates for younger offenders. It may well be the case that the older a recidivist offender gets, the more likely they have experienced ETS or a similar intervention, and the less effective a repeat experience will be. The ETS offenders who did reconvict had, on average, higher OGRS scores than those who did not, but did not differ in age from offenders who did not reconvict. This again highlights the need for correct targeting of offenders for ETS: nearly half (48%) of the cases who did reconvict had OGRS scores of 75 and over, whereas of those who did not reconvict only 16% had high OGRS scores. Given that offenders who complete the programme have much lower reconviction rates, it would be useful to see if the effects of age and risk upon reconviction found in this study on all offenders sentenced are also found for completers and non-completers. Much larger samples of completers than obtained in this study would be necessary to run reliable statistical analyses. The ETS offenders had lower actual reconviction rates for both men and women. These differences were too small to be found statistically significant with these samples and may have arisen due to chance. The difference between the comparison and the ETS samples’ actual reconviction rates is larger for the men: -7.6% compared to -2.1% for the women. When actual reconviction rates are compared against the predicted rates, it appears that only men in the ETS sample reconvict less than expected. Much caution is needed with the difference for the women due to the low numbers of women in the samples overall but this might suggest that although receiving ETS is better than not for both sexes, the programme in West Yorkshire is less effective with women than with men. Further research with larger samples of women is needed. The average on-set time to the first reoffence was 189 days for the ETS sample and 233 days for the comparison sample. This difference was not statistically significant suggesting no real difference in how quickly the two samples began reoffending again. However, during the first year the comparison group’s reconviction rate was lower than the ETS group’s rate, at a maximum of 6.5% lower 5 months after being sentenced. This trend reversed at approximately 8-9 months when the ETS group’s reconviction rate began to slow, falling below the comparison group’s rate after one year. By 18 months the ETS group’s reconviction rate was at its maximum difference of 6.6% lower than the comparison group. Poor targeting may in part cause this: offenders with higher OGRS scores reoffend faster in both samples and the ETS sample had higher proportions of these offenders. However, we can also theorise that the decrease in reoffending rate at the 8-9 month point is linked to offenders completing the programme, as the average time to completion was 7 months and offenders who complete have lower reconviction rates than predicted. A criticism of reconviction studies is that just one reconviction is necessary, perhaps for a relatively minor offence, for the offender to be counted as ‘reconvicted’. Considering the volume or frequency of reconvictions over a two-year period and the seriousness of these provides another more sensitive measure of reconviction. To look at frequency and seriousness however, each offender needs to have an equivalent period of two-years at liberty to reoffend, i.e. not in prison. Due to the time-frame of this study, this resulted in 90 offenders being excluded from frequency and seriousness analyses as they were in custody. Cases ‘lost to custody’ – 23% (n=70) from the ETS sample and 12% (n=21) from the comparison group – had higher OGRS scores than the remaining cases: 11% and 12% points for the comparison and ETS samples respectively. This suggests that the remaining ETS sample used in this part of the study is slightly skewed towards medium-high risk offenders and actual figures should be used with much caution. However, as the difference in OGRS scores between cases lost to custody and the final sample was proportionately the same the comparison between the two samples is not necessarily jeopardised by the exclusion of these cases. The ETS sample was responsible for more reconvictions and offences, on average per reconvicted offender. The ETS sample had an average of 2.26 reconvictions, compared to 2.05 for the comparison sample, although this difference is not sufficient to be statistically significant. The comparison sample also had a higher proportion of offenders with only 1 reconviction, this being 54.1% of those who reconvicted, compared to 41.5% in the ETS sample. It might be the case that more offences were committed by the ETS sample but done so by a smaller number of offenders. However this conjecture cannot be statistically confirmed or refuted due to the smaller number of offenders who actually did reconvict. For the purpose of this study, a quick and systematic method was used to gauge the seriousness of a reconviction using the courts’ view of seriousness as determined by the disposal given. This enabled a robust comparison between the two samples; it was not intended to make a theoretical point about offence seriousness per se. Approximately 39% of both samples’ reoffences were more serious than the index offence, whilst the ETS group had slightly higher proportions (+6.5%) of less serious reoffences than the comparison group. The comparison group had similarly higher proportions of same-seriousness reoffences. This implies that the reoffending committed by the ETS sample is no worse than that of the comparison group.

Details: Wakefield, UK: West Yorkshire Probation Service, 2006. 40p.

Source: Internet Resource: Accessed April 5, 2013 at: http://www.westyorksprobation.org.uk/documentlist.php?type=1&year=2006

Year: 2006

Country: United Kingdom

URL: http://www.westyorksprobation.org.uk/documentlist.php?type=1&year=2006

Shelf Number: 128272

Keywords:
Community-Based Corrections Cognitive Skills
Enhanced Thinking Skills
Probationers
Recidivism
Reoffending (U.K.)