Dynamic risk factors and treatment change : exploring the mechanisms of sexual offending onset and desistance
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The concept of change and the ability for individuals to reduce their level of risk through targeted intervention is a core feature of current rehabilitation frameworks used with individuals who have engaged in sexually harmful behaviours. However, despite the large emphasis placed on dynamic risk factors and the acknowledgement of the ability for individuals to make prosocial change, relatively little attention is given to furthering our understanding of how dynamic risk meaningfully relates to the aetiology of sexual offending, and to developing theories of the mechanisms underlying the change process. The current thesis therefore addressed this gap in the literature by investigating the factors and characteristics that play a causal role in sexual offending behaviour, and by exploring the underlying nature of offender change to help inform ongoing theory generation in this area.
Study One began with a validation of an influential theory of the aetiology of sexual offending, Ward and Siegert’s (2002) Pathways Model of Child Sexual Offending. The study used pre-treatment scores on a psychometric battery completed by 1,134 male sexual offenders against children to conduct a Latent Profile Analysis (LPA), which is a statistical technique used to identify meaningful latent classes of individuals within a given sample. Results suggested that the sample was best captured by five classes of individuals that mapped closely to the five hypothesised pathways in the Pathways model, with a few notable exceptions. Overall, the study provided tentative support for the Pathways Model and its proposed mechanisms and aetiological pathways, provided guidance for potential amendments to the model, and highlighted the heterogeneity in the offender population and causes of offending.
Studies Two, Three and Four then went on to explore the nature and characteristics of sexual offender change, with the aim of providing valuable insights to inform ongoing theory generation regarding the mechanisms and nature of change. Study Two provided the first known study to explore whether sexual offender treatment change is best conceptualised as categorical or dimensional, by using standardised residual change scores from 346 male sexual offenders against children to conduct a taxometric analysis of change. Results from the analysis suggested that offender change is best conceptualised as a categorical construct; that is, that differences in treatment change between individuals are best understood as differences in the types of change made, rather than simply the amount of change made.
Study Three explored the implications of Study Two’s findings further by attempting to identify what these change categories might look like. The study used standardised residual change scores from 1,170 sexual offenders against children to conduct an LPA, which found that three classes provided a best fit for the data. These classes represented individuals who had made Poor Change, Moderate Change, or Good Change over the course of treatment, with individuals in the Good Change group reoffending at significantly lower rates than individuals in the other two groups. The study suggested that meaningful distinctions can be made between different kinds of change made over the course of treatment, but did not provide much information regarding the mechanisms underlying these change patterns.
Study Four therefore provided a further investigation of these groups, by assessing the pre-treatment needs, static risk, and historical or demographic characteristics associated with each change group. Results indicated that there were no significant differences in static risk or most historic or demographic factors between groups, but that individuals in the Good Change group showed significantly lower rates of pre-treatment needs than individuals in the Poor Change group (with individuals in the Moderate Change group falling in between). This suggested that perhaps individuals in the Good Change group were already on a pathway to desistance prior to entering treatment.
Together, the results from this thesis suggest that internal factors, such as motivation to change and cognitive transformation (i.e. the adoption of a pro-social identity), may be key mechanisms underlying change demonstrated by offenders. They also highlight the heterogeneity of pathways into sexual offending and related treatment needs, and add to a growing body of research supporting the need for individualised assessment and intervention that focusses on the promotion of prosocial identity and skill acquisition.