Rhodes, Hayley M
(2023)
An Evidence-based Approach to Prioritising Indecent Images of Children Offenders: Implications for Deployment and End User Operationalisation.
Doctor of Philosophy thesis, University of Liverpool.
|
Text
200899461_Jun2023.pdf - Author Accepted Manuscript Download (1MB) | Preview |
Abstract
The proliferation of indecent images of children (IIOC) on the Internet and associated offending has created a growing workload for law enforcement globally, who have struggled to cope with the increasing investigative demand (McGuire & Dowling, 2013). The Kent Internet Risk Assessment Tool (KIRAT) was developed in response to this issue (Long et al., 2016). The tool aims to identify those suspected IIOC offenders most likely to have also engaged in previous and/or concurrent contact sexual offending against children for prioritisation. The tool also supports efficient resource management. This thesis investigates three KIRAT related areas; (i) development and cross-validation of KIRAT for international use, (ii) the effectiveness of the KIRAT online training, and (iii) exploring how KIRAT works, effects of its use and whether it’s effective in practice. These studies all contribute original knowledge. Development of the KIRAT Europe (EU) model resulted from the first study. The sample comprised of 1,148 male IIOC offenders (501 ‘higher risk of contact’ [HR] and 647 ‘lower risk of contact’ [LR]) based within 12 countries from around the world. Tests revealed highly significant differences in risk classification between dual and IIOC-only offenders from European and international countries. The removal of two items (past convictions/allegations for contact offending) also resulted in a statistically significant model, demonstrating the contribution of other criminal history and dynamic factors to risk prediction. The second study, a mixed methods evaluation, demonstrated that training completion did not produce increased accuracy in the identification of contact offending risk predictors post-training versus baseline. Practitioners performed particularly well at baseline however. Over 97% of all practitioners, and 98% of the novice-only group (no previous experience of KIRAT), passed the course on the first attempt. Practitioner demographic factors were not predictors of change in risk predictor error scores and were not significantly associated with passing or failing the course assessment. The sample was however, saturated with practitioners that had previous KIRAT experience. Intervention effects contributing to successful outcomes were; the comprehensive/interactive nature of the training, training structure/format, and the online format of the course. Factors that could hinger positive outcomes were also identified and require further consideration. Results of the third research study, a process and impact evaluation of KIRAT, revealed four intervention effects contributing to outcomes of KIRAT; action/response framework, tool simplicity, practitioner discretion, and minimum standard of investigation. Impacts of the tool include reducing victimisation and the action/response framework. Future economic evaluation to model cost-benefits of KIRAT and suggestions to address the gap in meeting police enforcement timescales were made. The findings reveal that KIRAT EU, and the online training, could potentially be used by police agencies worldwide to safeguard children from harm. The tool has been demonstrated as fit for purpose and the training an effective measure to provide the requisite knowledge to deploy KIRAT in police investigations. Within the Discussion, research implications, study limitations and future research are considered.
| Item Type: | Thesis (Doctor of Philosophy) |
|---|---|
| Divisions: | Faculty of Health and Life Sciences Faculty of Health and Life Sciences > Institute of Population Health |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 05 Feb 2024 16:52 |
| Last Modified: | 05 Feb 2024 16:52 |
| DOI: | 10.17638/03176859 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3176859 |
Dimensions
Dimensions
