Understanding the Effect of Immunity on Over-Dispersed Criminal Victimizations: Zero-Inflated Analysis of Household Victimizations in the NCVS

Abstract

This study aims to empirically test the immunity effect on the frequency distribution of household victimizations. To clarify the immunity effect, the statistical construction of zero-inflated models is reviewed and compared with that of non-zero-inflated models. The Benjamini and Hochberg correction is used to address the limitation of p values in multiple testing. Compared with the findings from the non-zero-inflated model, two sets of coefficients from the zero-inflated model reveal that there exist more complex and diverse statuses in the process of household victimization than predicted by risk heterogeneity and event dependence. With these findings, this study suggests that zero-inflated models should be introduced and compared with non-zero-inflated models for the clarification of victimization determinants.

Immigrant status, citizenship, and victimization risk in the United States: New findings from the National Crime Victimization Survey (NCVS)*

“…Both tables contain three models, which correspond to the hypotheses H1, H2, and H3, respectively. We find that the control variables exhibit patterns that are highly similar to those reported in several previous studies of victimization risk based on the NCVS (e.g., Lauritsen & Carbone-Lopez, 2011;Park & Fisher, 2017;Rezey, 2020;Xie & Baumer, 2018). For example, as the results for H1 in both tables 5 and 6 indicate, having higher income, owning a home, being older, being married, living outside of central cities, and living in the Northeast are all related to lower risks of violent and property victimizations.…”

The Effects of Life Domains on Cyberbullying and Bullying: Testing the Generalizability of Agnew’s Integrated General Theory

“…7.3). To determine the most adequate Poisson-based regression model for a dependent variable of interest, the mean and the variance of the Poisson distribution was examined using R. The result indicated that the frequency distribution of cyberbullying was more over-dispersed than expected by a random and independent Poisson distribution (Park & Fisher, 2017). This study, therefore, calculated negative binomial regression models to consider that the expected variances of cyberbullying exceeded the variance of the Poisson distribution.…”

Untangling the Complex Pathways to Confidence in the Police in South Korea: a Stepwise Multilevel Structural Equation Modeling Analysis

“…When γ is given as 2, this function becomes the AIC statistics. If γ is equal to log(N), this function is identified as the BIC statistics, and if γ is identical to log((N + 2)/24), this function is equal to the ABIC statistics (Muthén & Muthén, 1998Park & Fisher, 2017). As these model-fit statistics control for the number of parameters, they can compare both nested and non-nested models (Dayton, 2003).…”