Individual Submission Summary
Share...

Direct link:

Who You Know or How You Know Them: Using a Decade of Web Scraped Arrest and Victimization Data to Examine the Influence of Co-offending Network Position on Victimization

Wed, Nov 14, 5:00 to 6:20pm, Marriott, International 3, International Level

Abstract

Co-offending is a precarious social interaction, requiring risky engagement with other offenders. However, the precariousness of these interactions is likely dependent on subjects’ position within their criminal network. The current study examines the relationship between subjects’ position in co-offending networks and the number of experienced victimizations. This project uses 10 years of web scraped arrest and victimization data made available online by many US Police Departments via Police2Citizen Daily Bulletin boards. As far as we are aware we are the first to do so. Negative binomial regression is used to examine covariance between network centrality and the number of victimization for each subject. Preliminary findings indicate that, net of the number of prior arrests, subjects with greater closeness centrality are victimized more frequently. Conversely, subjects with greater degree, betweenness, and eigenvector centrality are victimized less frequently. This suggests that, net of the positive association between arrests and victimizations, engaging in crime with (i) a greater number of other offenders, (ii) more than one group of offenders (e.g. gangs), and (iii) offenders who have co-offended with a greater number of other offenders are protective. Conversely, subjects who are close to every member of the co-offending network are at greater risk.

Authors