Abstract
Background: Extensive research has examined the impact of staff attitudes on the rehabilitation of men who commit sexual offences. However, investigations of automatic attitudes towards men with IDD detained within forensic mental health services with a history of committing sexual offences are limited. The evaluation of non-automatic attitudes continues to dominate this field, with insufficient consideration given to attitude models in developing staff educational workshops. The Motivation Opportunity Determinants processing (MODE) and Associative-Propositional Evaluation (APE) attitude models formed the theoretical basis for the workshop used in this study. This thesis aimed to examine staff’s automatic and non-automatic attitudes and perceptions towards men with IDD who commit sexual offences, and those who commit non-sexual violent offences in order to develop a deeper understanding of attitudes towards complex populations.Method: Quantitative analyses evaluated data collected from eight forensic mental health services and a university within South Wales. Automatic and non-automatic attitudes were assessed by those who did or did not participate in an intervention (a three-day workshop). Direct measures included a Brief Knowledge Survey (developed by the author), the Working with Offenders Questionnaire (WOQ), and three validated measures: the Attitudes Towards Prisoners (ATP), Attitudes Towards Sexual Offenders (ATS) and the Community Attitudes Towards Sexual Offenders (CATSO). The indirect measure included a Multi-Dimensional Implicit Association Test (IAT), (computer and pencil and paper named version) developed by the author. Data collection occurred immediately prior to and post-completion and three months after the workshop to investigate the maintenance of any attitude change.
Results: The direct measures data from all participants demonstrated that non-automatic attitudes towards men who commit sexual offences were more positive than anticipated, and increases were generally maintained over time. Significant differences between the intervention and non-intervention samples for the ATP and ATS were reported. Higher levels of positive attitudes were found for the non-intervention group for the ATS than the intervention sample, although positive increases were demonstrated post-workshop. The indirect measures data indicated automatic attitudes were more positively associated with people who have not committed violent offences when compared to individuals who have. Significant differences were found for the student (non-intervention) sample towards individuals who have committed sexual offences when compared to individuals who committed violent offences. The opposite outcome for the intervention sample (staff and students) was found. Further evaluation of the intervention sample demonstrated significant differences. The staff demonstrated more positive associations towards those who had committed sexual offences than individuals with violent offences, whereas the student sample demonstrated the opposite direction of bias. Additionally, the intervention sample demonstrated more positive associations towards individuals with IDD when compared to those who had committed sexual offences. In contrast, staff responses were less indicative of an attitude bias. Reliability and validity tests for the computer Md-IAT demonstrated ‘good to excellent’ levels of internal consistency and test-retest outcomes. However, no significant correlations were found from the construct validity tests. Thus, results from the Md-IAT cannot be interpreted confidently. For the pencil and paper version, levels of internal consistency greatly varied and reflected other research findings. Therefore, caution needs to be applied when interpreting the findings.
Conclusions: Staff levels of positive attitudes towards people who commit sexual offences were high prior to the workshop and increased further upon completion, possibly indicating an influence of ‘contact’ and attendance at the workshop. Contribution to knowledge: The study adds to the existing body of knowledge by evaluating automatic attitudes within forensic settings using a multiple-target IAT. It helped identify unexpected directions of bias in automatic attitudes, which a single-target IAT would not have. Findings provide a deeper knowledge base of staff attitudes when working with men who commit sexual offences and other complex presentations. A potential benefit of examining automatic attitudes is that it can guide staff support, enhance the rehabilitative environment, and use staff workshops with an attitude model as a theoretical basis. However, this approach requires further investigation, with a standardized method to test validity of the Md-IAT.
Date of Award | 2024 |
---|---|
Original language | English |
Supervisor | Rachel Taylor (Supervisor) & Philip Tyson (Supervisor) |