Varity and Frequency Scales of Antisocial Involvement: Which One is Better?

• NCJ Number: 202683
• Journal: Legal and Criminological Psychology Volume: 8 Issue: 2 Dated: September 2003 Pages: 135-150
• Author(s): Mons Bendixen; Inger M. Endresen; Dan Olweus
• Date Published: September 2003
• Length: 16 pages
• Publication Format: Article
• Publication Type: Research (Applied/Empirical)
• Language: English
• Country: United Kingdom

Annotation

This study examined to what extent a variety scale of antisocial behavior would benefit from the inclusion of frequency scores of antisocial acts with regard to internal consistency, stability, and associations with conceptually related variables.

Abstract

A variety scale is a measure reflecting the range or number of diverse antisocial acts the respondent has admitted to in a specified period. This scale is typically regarded as a good indicator of an individual’s involvement in antisocial behavior. This study examined whether a scale including frequency scores of antisocial behavior was a more sensitive and better measure of antisocial involvement than a variety scale. The study is part of a large-scale longitudinal research program. Data were collected from four relatively large cohorts of students from the city of Bergen, Norway. The sample consisted of 1,292 Norwegian student aged 13 and 14 years. Data from this sample were used to compare a 17-item variety scale with 2 versions of a frequency scale covering the same 17 items. The study examined the internal consistency of the scales, stability coeffiecients, for 1-year and 2-year intervals, and associations with continually related variables. The results indicate that the (raw) frequency scale was not a more reliable or valid indicator of a student’s antisocial involvement than the variety scale. All the analyses yielded poorer results for the frequency scale than for the variety scale. In conclusion, the results argue against replacing a variety scale with a frequency scale. A variety scale has the advantage of being easy to interpret theoretically, as well as the mean difference or an unstandarized regression coefficient can be given a quite concrete and specific meaning. However, there may be some minor benefits from using an appropriately transformed frequency scale as a supplement to a variety scale. Tables and references