Does change in psychometric test scores tell us anything about risk of reconviction in sexual offenders?

• Published in: Psychology, crime & law crime & law, 2013-01, Vol.19 (1), p.85-110
• Main Authors: Barnett, Georgia D., Wakeling, Helen, Mandeville-Norden, Rebecca, Rakestrow, Janine
• Format: Article
• Language: English
• Subjects: Adult and adolescent clinical studies; Biological and medical sciences; Medical sciences; Medical treatment; Mental health care; Methodology. Experimentation; Offenders; Psychology. Psychoanalysis. Psychiatry; psychometric test change; Psychometrics; Psychopathology. Psychiatry; Quantitative psychology; Recidivism; reconviction; Reliability; Risk; Risk assessment; Sex crime; Sex offenders; Sexual behavior disorders. Psychogenic sexual dysfunctions; sexual offenders; Techniques and methods; treatment; Violent crime
• ISSN: 1068-316X; 1477-2744
• DOI: 10.1080/1068316X.2011.607820

This study examined the relationship between psychometric test change over treatment and sexual or violent reconviction. A sample of 3402 convicted sexual offenders attending probation sexual offender treatment in the community completed psychometric tests pre- and post-treatment. Individuals' treatment change status was determined by whether their pre- to post-treatment scores had reliably deteriorated, were unchanged, reliably improved (but not to within a functional range), reliably improved to, or remained within, a functional range. Treatment outcome status was not associated with reconviction for a sexual or violent reoffence for any of the measures examined. The sample was then grouped into those who had and had not changed to within a functional range of scoring on amalgamated measures of offence-supportive attitudes, socio-affective functioning and self-management. Those who were classed as not requiring further change on the socio-affective functioning domain had a lower rate of reconviction than those who, according to the psychometric tests, still needed to change post-treatment. However, change on this domain did not add predictive power to static risk assessment. Overall treatment change was not associated with reduced sexual or violent recidivism. Recommendations for further research in this area, and the practical implications of these findings, are discussed.