The Maudsley Violence Questionnaire: initial validation and reliability

• Published in: Personality and individual differences Vol. 38; no. 1; pp. 187 - 201
• Main Author: Walker, Julian S.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 2005; Elsevier
• Subjects: Beliefs; Biological and medical sciences; Cognition; Cognitions; Delinquency; Machismo; Medical sciences; Psychology. Psychoanalysis. Psychiatry; Psychometrics. Diagnostic aid systems; Psychopathology. Psychiatry; Questionnaire; Techniques and methods; Violence; Questionnaire; Machismo; Violence; Cognitions; Beliefs; Delinquency; Human; Test validation; Factor analysis; Psychometrics; Test reliability; Social behavior disorder
• ISSN: 0191-8869; 1873-3549
• DOI: 10.1016/j.paid.2004.04.001

This paper describes the construction of the Maudsley Violence Questionnaire (MVQ) and the investigation of the factor structure and reliability. The MVQ measures a range of cognitions, relating to violent behaviour, drawn from clinical and theoretical perspectives, but with justification of violence in response to threatened self esteem and the legitimising of violence as central elements. 785 students (male and female) between 16 and 19 years of age, from a range of backgrounds were assessed using the MVQ and a Self Report Delinquency Scale (adapted from Mak, 1993). A principal axis factoring analysis was conducted and two factors were rotated using direct oblimin procedure. Although the separate factor analyses for men and women generated the same factors, the respective factor loadings differed slightly. The male factor structure was used given that men are responsible for the majority of violence and are likely to comprise the majority of respondents for the MVQ in forensic clinical practice. The scale comprised two factors: ‘machismo’ (42 items) and ‘acceptance’ of violence (14 items); the Cronbach alpha coefficients ranged from 0.74 to 0.91; and, t-tests revealed a significant difference between men and women on both the factor total scores. In accordance with predictions, self-reported violence in men was predicted by both machismo and acceptance (most strongly by machismo); however, for women self-reported violence was predicted more strongly by acceptance than machismo.