ORCID

https://orcid.org/0000-0001-8788-7182

Date of Award

Summer 2026

Language

English

Embargo Period

7-7-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

School of Criminal Justice

Program

Criminal Justice

First Advisor

Matt Vogel

Second Advisor

Shiguo Jiang

Committee Members

Martin Andresen; David Hureau; Robert Worden

Keywords

Crime Concentration; Near Repeat Victimization; Random Effects; Rish Homogeneity Hypothesis

Subject Categories

Criminology and Criminal Justice

Abstract

Studies of crime and place have long asked how much random processes contribute to the spatial and spatiotemporal concentration of crime. Traditionally, simulations assume that crime incidents occur randomly across microgeographic units—such as block groups, street segments, and parcels—by selecting geographic units at random with replacement. This dissertation takes a different approach, modeling crime incidents as random selections among potential targets, also with replacement. In this framework focusing primarily on burglary, simulated patterns reflect the spatial distribution of potential targets (i.e., properties). This dissertation relies on burglary data from San Antonio, Texas (2017–2019), and measures available burglary targets as individual physical addresses. Under this framework, this study estimates random crime concentration, defined as the stochastic component of crime concentration due to random processes. Here, random crime concentration is conceptualized as formed by burglary incidents occurring randomly to all physical addresses nested within geographic units. By refining the conceptualization of random crime concentration, this research offers a clearer understanding of the role of random processes in crime concentration and the influence of target availability on observed patterns than prior research. This dissertation also considers Ripley’s K-function and its derivatives as alternative measures for spatial and temporal concentration of crime, examining whether this incident-level, distance-based approach outperforms common methods of measuring concentration, such as Weisburd’s metric, the (generalized) Gini coefficient, or the Knox test. My dissertation comprises two distinct but interrelated studies. The first examines the random effects in spatial concentration of crime (i.e., crime concentration at place). The second considers the random near repeat effects, the stochastic component of spatiotemporal concentration of crime (i.e., near-repeat victimizations). I further generalize the conceptualization of random crime concentration in the spatiotemporal context, considering crime incidents as occurring randomly to targets at random time points. The results suggest that (1) random effects cannot explain either spatial or spatiotemporal concentration of crime entirely; (2) random concentration of burglary can account for approximately half of the observed concentration of burglary; and (3) the global results are driven by a small group of places with highly concentrated burglary incidents. These findings have implications for crime prevention policy and suggest new directions for research on crime and place.

License

This work is licensed under the University at Albany Standard Author Agreement.

Download

Share

COinS