Abstract
In this study, the internal reliability and construct validity of the recently adapted Swedish version of the Novaco Anger Scale (NAS-1998-S; Lindqvist, Dåderman, & Hellström, Social Behavior and Personality, 8, 773–788, 2003), as well as its scale correlations with demographic and criminality variables, were investigated. Construct validity was established by assessing the correlation pattern of the scales of NAS-1998-S with concurrent scales of similar and distinct constructs. Ninety-five male violent prisoners, ranging in age from 18 to 67 years, participated. The results demonstrated good internal reliability, consistent intrascale relationships, and appropriate construct validity of NAS-1998-S. The number of previous convictions had a moderate negative relationship with the capacity of control. Age and education correlated negatively with the NAS-1998-S scales, except Regulation. In addition to psychometric issues, the results were discussed from a clinical perspective on the offender population.
As a part of a doctoral dissertation by Judit Krisztina Lindqvist, this study has been supported and financed by a doctoral fellowship from the Department of Psychology, Stockholm University. The authors gratefully acknowledge Lars Krantz, Kenneth Gustafsson, Hans-Olof Larsson and Olof Vannäs at the Swedish Prison and Probation Service for their assistance and support in the selection of participants and collection of data in this study.
Notes
1 The choice of violent convictions, exclusively, was due to the purpose of the research project (the one that this study is a part of), namely, the comparison of two samples that differed with regard to their aggressiveness, operationalized as convictions for physically violent crimes.