Evaluating the factor structure and reliability of the Community Attitudes Toward Sex Offenders (CATSO) scale

Purpose - The study explored the factor structure and reliability of the Community Attitudes Toward Sex Offenders (CATSO) scale. The study also included an impression management scale. Design/methodology/approach - A total of 441 adults (134 male, 280 female, 27 ns) participated in this study from three populations: staff at a category B prison (n = 62), staff at a category C sex offender prison (n = 102) and staff at a UK university (n = 248). Questionnaire packs included information/consent, demographics, the CATSO and the Paulhus impression management scale. Findings - Data were excluded where the Paulhus score was <1 or >12 (faking good/bad present). Confirmatory factor analysis with alternative models indicated the scale did not meet any of the requirements for an acceptable fit. Cronbach's a confirmed that two of the four sub-scales were not internally consistent. An exploratory factor analysis was conducted following the removal of items with poor item-total correlation and/or low/high facility index and, following parallel analysis, a revised two factor solution was examined. The CATSO needs revision; it is unclear whether it is sufficiently reliable and valid for use in the UK. The need for a valid/reliable tool to assess attitudes toward sexual offenders remains an important goal for researchers. Originality/value - Church et al. (2008) developed a scale (CATSO) to measure attitudes toward sex offenders; the scale is being increasingly widely used across a range of populations, including the general public and correctional staff. This research identifies significant problems with the scale in terms of factor structure and reliability of the sub-scales. This paper advises a rethink of the CATSO by the scale authors and suggest the scale is not currently useable.