Exploring the Characteristics of Personal Victims Using the National Crime Victimization Survey

摘要

This thesis endeavors to investigate the effects of different individual-level and household-level characteristics (such as marital status, employment, and active lifestyle) on personal victimization prevalence, incidence and concentration. The main attributes of interest explored in this analysis targets vulnerability, opportunity, attractiveness, and area population. Two statistical methods were used to investigate these hypotheses, logistical regression for victimization prevalence, and negative binomial regression for victimization incidence and concentration. Results were in accord with empirical expectations of all hypotheses. The evidence clearly suggests that there are individual-level and household-level characteristics that influence prevalence, incidence, and concentration. Evidence from the model of concentration implies that although there are characteristics that are shared by victims, not all characteristics that influence one's chances of becoming a victim seem significant for repeat victims. Further the measure of overdispersion from the negative binomial analysis of concentration suggests that there was very little unexplained heterogeneity when compared to the analysis of incidence. Hence the statistically significant explanatory variables in this model accurately captures the characteristics shared by many repeat personal victims.