An Investigation Into the Factor Structure of the Attitudes to Suicide Prevention Scale

• Published in: Crisis : the journal of crisis intervention and suicide prevention 2020-03, Vol.41 (2), p.97-104
• Main Authors: Sandford, David M., Kirtley, Olivia J., Lafit, Ginette, Thwaites, Richard, O'Connor, Rory C.
• Format: Article
• Language: English
• Subjects: Administrative Personnel; Attitude Measures; Attitude of Health Personnel; Attitude to Health; Between Groups Design; Community Health Services; Community Mental Health Services; Confirmatory Factor Analysis; Exploratory Factor Analysis; Factor Analysis, Statistical; Factor Structure; Female; Health Personnel; Human; Humans; Male; Professional Role; Psychometrics; State Medicine; Suicide - prevention & control; Suicide Prevention; Surveys and Questionnaires; Training; United Kingdom
• ISSN: 0227-5910; 2151-2396
• DOI: 10.1027/0227-5910/a000608

Aim: The aim of this study was to investigate the factor structure of the Attitudes to Suicide Prevention Scale (ASPS). Method: The ASPS was distributed to all staff in a UK National Health Service Trust (N = 957). We conducted an exploratory factor analysis followed by a confirmatory factor analysis by splitting the data 60/40 into training and testing subsets. A multiple regression analysis was carried out to investigate whether the overall scale score varied as a function of professional role, age, and gender and whether respondents had completed suicide prevention training or not. Results: Two items displaying poor item-scale correlation were excluded from the factor analysis and a further item was excluded as it was based on different anchor points. For the remaining 11 items, no adequate factor structure emerged. The scale total demonstrated statistically significant differences in attitudes between staff groups (defined by attendance at suicide awareness or prevention training, by gender, and by level of patient contact), but not between groups defined by age range. Generally, however, there were positive attitudes across all Trust staff. Limitations: This study had a low response rate (24%) and was cross-sectional which limits the conclusions that could be drawn. Furthermore, other areas such as convergent validity and test-retest reliability were not examined. Conclusion: Our findings found no satisfactory factor structure for the ASPS. Further scale development would be beneficial.