Citation

Tseloni A, Pease K. Br. J. Criminol. 2004; 44(6): 931-945.

Copyright

(Copyright © 2004, Centre for Crime and Justice Studies, Publisher Oxford University Press)

DOI

10.1093/bjc/azh047

PMID

unavailable

Abstract

The paper seeks to clarify the risks of repeat personal victimization, acknowledging that much prior research is limited in enabling efficient applicability to crime-prevention purposes by its failure to address the interplay between personal and household factors conferring risk alongside the effects of prior victimization. The study reported here is an attempt to pin down sources of unexplained heterogeneity in terms of random effects of known characteristics of individuals and their households, their lifestyle and victimization history. The data for this study came from the 1994 US National Crime Victimization Survey, and focuses upon personal victimization. Individual, household and prior non-victimization fixed effects are found to be highly statistically significant, implying that each set of covariates adds important information for the prediction of personal crimes. Recent (6 < 12 months) event dependence effects are found to vary according to lifestyle and area of residence. Adding random effects decreased the estimated between-individual variation of the mean number of personal crimes by roughly a third over and above fixed effects. In principle, the data here could provide police and their partners engaged in crime reduction with simple but systematic notions of who is at most risk beyond the short term, applicable to every case, incorporating information about households and individuals which includes prior victimization, and allow them to advocate measures of self-protection commensurate with such risk. Such differential treatment of households similarly placed in terms of prior victimization should have its ethical implications discussed, and further research which unpacks' personal crime into its constituent offences seems necessary.