Measuring Online–Offline Spillover of Gang Violence Using Bivariate Hawkes Processes

ObjectivesThe goal of this research is to investigate the association between online and offline gang conflict. It does this by investigating the magnitude and causal ordering of this association with a case study of Chicago Latino gangs.MethodsChicago Police Department records of gang shootings (N...

Ausführliche Beschreibung

Gespeichert in:  
Bibliographische Detailangaben
1. VerfasserIn: Leverso, John (VerfasserIn)
Beteiligte: Diouane, Youness ; Mohler, George
Medienart: Elektronisch Aufsatz
Sprache:Englisch
Veröffentlicht: 2025
In: Journal of quantitative criminology
Jahr: 2025, Band: 41, Heft: 1, Seiten: 103-131
Online-Zugang: Volltext (lizenzpflichtig)
Journals Online & Print:
Lade...
Verfügbarkeit prüfen: HBZ Gateway
Schlagwörter:
Bivariate Hawkes process
Criminal Behavior
Gang violence
Gangs
Quantitative Criminology
Social Media
Sociological Theory
Sociology
Statistics
Beschreibung
Zusammenfassung:ObjectivesThe goal of this research is to investigate the association between online and offline gang conflict. It does this by investigating the magnitude and causal ordering of this association with a case study of Chicago Latino gangs.MethodsChicago Police Department records of gang shootings (N = 566) are combined with 9873 gang confrontations on social media from a Facebook page devoted to Chicago Latino gangs. A bivariate Hawkes point process is then fit to the data to estimate spillover effects from Facebook to Chicago street violence and vice versa.ResultsWe estimate that each shooting causes 0.068 (0.015, 0.182) negative Facebook comments directed towards the victim gang. We estimate that each negative Facebook comment directly causes 0.002 excess shootings, though this effect is not statistically significant. When focusing on the three most active gangs, the measured spillover effects are even larger. We estimate for the three most active gangs, 9% of Facebook comments are caused by shootings and 3% of shootings are caused by negative Facebook comments, however the latter effect is not statistically significant.ConclusionsThe data indicates that most online negative interactions between gangs stay online. Further we find that the only causal relationship the data supports is that offline violence leads to negative interactions online. We did not find statistically significant evidence of a causal relationship in which online interactions lead to offline violence. Finally, the data suggests contextual considerations, such as the size of the gang, need to be considered when assessing such relationships.
ISSN:1573-7799
DOI:10.1007/s10940-024-09592-5