Could Social Medias Reflect Acquisitive Crime Patterns in London?: —-Insights Derived from Broken Windows Theory

TY - JOUR

T1 - Could Social Medias Reflect Acquisitive Crime Patterns in London?: —-Insights Derived from Broken Windows Theory

AU - Wang, Zenghui

AU - Li, Yijing

PY - 2022/6

Y1 - 2022/6

N2 - Embraced within the framework of crime opportunities integrated with Social Disorganization theory and Broken Windows theory, this paper intends to explore the patterns of four types of acquisitive crimes, using social media data i.e. Twitter, Foursquare and cross-sectional data acquired through text analysis technique. With Greater London as the study area, models like negative binominal regression (NBR) and geographically weighted regression (GWR) are performed to illustrate the aggregated relationships between acquisitive crimes and crime opportunities at London-wide and sub-regional MSOAs levels respectively. The results work towards to hypotheses that: the tweets sentiment could reflect property-related crime rates positively in light of Broken Windows Theory; more tweets with negative sentiment may incur increases of acquisitive crimes. It contributed to existing studies in (1) providing empirical evidence for integrating these three theories; (2) complementing current research on local discrepancies of acquisitive crimes by utilising both GWR and NBR models; (3) challenging the traditional stereotypes about racial disparities with the finding that ethnic heterogeneity and instrumental crimes have counterintuitive association, especially taking education factor into consideration; (4) implicating some localised acquisitive crime prevention strategies to policy makers in light of the reality that the relationship between local variations and different crime types may vary by place.

AB - Embraced within the framework of crime opportunities integrated with Social Disorganization theory and Broken Windows theory, this paper intends to explore the patterns of four types of acquisitive crimes, using social media data i.e. Twitter, Foursquare and cross-sectional data acquired through text analysis technique. With Greater London as the study area, models like negative binominal regression (NBR) and geographically weighted regression (GWR) are performed to illustrate the aggregated relationships between acquisitive crimes and crime opportunities at London-wide and sub-regional MSOAs levels respectively. The results work towards to hypotheses that: the tweets sentiment could reflect property-related crime rates positively in light of Broken Windows Theory; more tweets with negative sentiment may incur increases of acquisitive crimes. It contributed to existing studies in (1) providing empirical evidence for integrating these three theories; (2) complementing current research on local discrepancies of acquisitive crimes by utilising both GWR and NBR models; (3) challenging the traditional stereotypes about racial disparities with the finding that ethnic heterogeneity and instrumental crimes have counterintuitive association, especially taking education factor into consideration; (4) implicating some localised acquisitive crime prevention strategies to policy makers in light of the reality that the relationship between local variations and different crime types may vary by place.

UR - http://www.scopus.com/inward/record.url?scp=85130317229&partnerID=8YFLogxK

U2 - 10.1016/j.jnlssr.2021.08.007

DO - 10.1016/j.jnlssr.2021.08.007

M3 - Article

VL - 3

SP - 115

EP - 127

JO - Journal of Safety Science and Resilience

JF - Journal of Safety Science and Resilience

IS - 2

ER -