NCJ Number
205979
Journal
International Review of Victimology Volume: 10 Issue: 3 Dated: 2004 Pages: 245-260
Editor(s)
David Miers
Date Published
2004
Length
16 pages
Annotation
This article assesses the complex measures of cross-cultural invariance and gender bias in measuring the fear of crime.
Abstract
Over recent years there has been a tendency towards using scaling techniques as a “better” measure of the complex and multidimensional concept of fear of crime. However, using more complex measures can lead to an increase in the number of new issues related to measurement error. This article specifically addresses cross-cultural measurement invariance and gender bias as potential dangers to measurement invariance. Cross-cultural measurement invariance is such an issue because in order for cross-cultural comparison of fear of crime to be meaningful the instruments used to measure the theoretical constructs, such as fear of crime have to exhibit adequate cross-cultural equivalence. Gender bias plays a significant part in fear of crime literature and has found and presented different fear of crime levels between men and women. Utilizing the 1999 General Election Study, a large general population survey for Belgium on a variety of topics (N=4,239), this article attempts to demonstrate the techniques which could be used to test other sets or scales aiming to measure fear of crime. Both cross-cultural invariance and gender equivalence are assessed. This offers an illustration of the assessment of measurement invariance in a set of fear of crime items across cultures, such as the Flemish and Walloons in Belgium and between men and women. In summation, the issue of measurement invariance and the techniques to assess it gain the attention and application deserved, within other research domains. References