U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Fear of Crime and Victimization: A Multivariate Multilevel Analysis of Competing Measurements

NCJ Number
European Journal of Criminology Volume: 5 Issue: 4 Dated: October 2008 Pages: 387-409
Andromachi Tseloni; Christina Zarafonitou
Date Published
October 2008
23 pages
Using data from Athens, Greece, this study simultaneously modeled three commonly used indicators of fear of crime against direct (being a victim) and indirect (knowing a victim) victimization, controlling for demographic and socioeconomic characteristics of individuals.
The study found that the three features of fear of crime--feeling unsafe alone at home after dark, feeling unsafe walking alone after dark, and worry about becoming a victim of crime--weakened as key explanatory factors for fear of crime were considered. Direct and indirect prior victimization predominantly shaped perceived future risk and emotional states and behaviors related to fear of crime. Further, direct and indirect experiences of crime were more directly linked to feeling unsafe walking alone after dark and worry about becoming a victim of crime. Feeling unsafe at home alone after dark stemmed from the other two features of fear of crime. This suggests that feeling unsafe at home alone after dark is not as helpful as the other two measures in surveys for determining the strength of a person’s fear of crime. The data for this study came from a survey on “Insecurity, Fear of Crime, and Attitudes Towards the Criminal Phenomenon,” which was conducted in the Greater Athens metropolitan area in the spring of 2004. The sample consisted of 450 respondents selected on the basis of residence through stratified sampling. The effects of indirect and direct victimization on the three fear-of-crime measures were jointly estimated using multivariate multilevel logit modeling. 5 tables, 1 figure, 48 references, and appendix supplementary information on the multivariate multilevel logit model and a table indicating the effects of individual characteristics on victimization risk during the previous year via logit modeling