An investigation of attrition from community-based offending behaviour programmes

Hatcher, Ruth
An investigation of attrition from community-based offending behaviour programmes. Doctor of Philosophy thesis, University of Liverpool.

[img] PDF
HatcherRuth_Apr_2009_1220.pdf - Author Accepted Manuscript
Available under License Creative Commons Attribution.

Download (2MB)


Objectives: This thesis investigates attrition from community-based cognitive-skills offending behaviour programmes. Part one of the thesis investigates the influence of attrition on the reconviction outcomes of those sentenced to either the Enhanced Thinking Skills (ETS) programme or the Think First programme. The data are investigated to determine whether programme dropout is detrimental to reconviction outcomes. Part two undertakes a more thorough analysis of the characteristics of programme completers, non-completers, and non-starters of the ETS programme. The three groups are compared to assess for differences in relation to demographic, psychometric, criminal history, and offender need variables. The role of organisational, or process, factors in attrition is also investigated. Finally, the reasons recorded in probation files for non-attendance at the ETS programme are examined. Methods: Part one comprises two chapters and utilises a quasi-experimental design. Data relating to a national sample of offenders sentenced to an offending behaviour programme and a comparison group of offenders sentenced to probation but not required to undertake a programme and matched on a one-to-one basis to the experimental group are utilised. Part two utilises data relating to offenders sentenced to the ETS programme within one probation area. The focus on one locality permitted a rich analysis of the factors associated with attrition. Analyses undertaken include tests of association (correlations, chi-square), parametric and non-parametric tests for differences (t-tests, ANOVAs, Kruskal Wallis, Mann Whitney), logistic regression (binary and multinomial), and calculations of effect sizes. Results: The analyses in part one provided tentative evidence of a negative impact of attrition on reconviction outcomes. This trend remained after controlling for those variables found to significantly differ between groups. The analyses within part two indicated that programme non-starters were more criminogenic than programme completers and non-completers. The non-completers, however, were the youngest of the groups and hence had less of a history but displayed a similar rate of offending as the non-starters. In relation to process factors, there was a significant association between appropriateness of targeting and attrition; offenders with risk of reconviction scores above the recommended criteria were most likely to dropout. Finally, a third of dropouts could not, should not, or were not able to attend due to the unavailability of a programme place, a further third were already in breach of their order or had committed a further offence, and the final third could and should attend, were not apparently in breach but still failed to commence. Conclusions: Programme dropouts produce worse reconviction outcomes than programme completers and matched comparisons. In evaluating correlates of attrition, dropouts are more criminogenic than programme completers. However, process factors were also associated with programme attrition. Research should investigate the impact of individual and process factors on attrition further and should use these findings to inform the debate concerning the influence of programme attrition (and hence completion) on reconviction outcomes. It is anticipated that the findings will inform clinical practice and the treatment readiness and intervention outcome research literatures.

Item Type: Thesis (Doctor of Philosophy)
Additional Information: Date: 2009-04 (completed)
Uncontrolled Keywords: Attrition, offending behaviour programmes, offenders, probation, evaluation
Subjects: ?? HV ??
?? BF ??
Divisions: Faculty of Health and Life Sciences > Institute of Population Health
Depositing User: Symplectic Admin
Date Deposited: 08 Dec 2010 12:30
Last Modified: 16 Dec 2022 04:33
DOI: 10.17638/00001220