Associations between Greenspace and street Crimes in Toronto: Evidence from a spatial analysis study at dissemination area level
Loading...
Date
2020-06-03
Authors
Onifade, Odunayo
Journal Title
Journal ISSN
Volume Title
Publisher
University of Waterloo
Abstract
Introduction: Earlier criminologists have explored various factors generating or attracting crime in urban cities coupled with crime studies focusing on the influence of social, built and natural environments in urban centres. According to Statistics Canada (2019), the Crime severity index of Canada and Toronto has been on the rise since 2014, which found the violent crime severity index
showing higher trends than non-violent crime severity. This study, first, examined the crime trends and seasonality in Toronto. Next, the association between greenspace variables and street crime rates across the city at the dissemination level using the spatial statistical methods were explored. Previous crime studies have also investigated the relationship between the crime rate (property and violent) and greenspace, albeit this study only focused on analyzing crime that usually occurs outsides, namely “street crimes.” There are two schools of thought concerning the association between crime rates and greenspace. The first belief suggests greenspace facilitates criminal activities because it conceals the offender from the victims/bystanders, while the second belief insists that greenspace deter criminal activities.
Methods: Street crime considered for this research included assault, auto-theft and robbery crime. This study explored the association between greenspace variables and street crime rates across the City of Toronto. Crime data were extracted from the Toronto Police Service public safety data portal; the greenspace data were obtained from Toronto Open Data, and sociodemographic data were exported from the 2016 Census Data collected from Statistic Canada. Street crime trends and seasonality were first, and they were carried out using crime data from the year 2014 to 2018 to generate a line graph depicting yearly, monthly and seasonal crime trends in the study area.
Greenspace variables (stem density; basal area density and tree density) were estimated from tree inventory data. The sociodemographic variables considered were median household income, lone parent, unemployment rate, high school degree holders, owner-occupied housing and renter-occupied housing. Spatial distribution maps for the dependent and independent variables were generated to show the geographical variation of the data. The Global Moran’s I and Local Indicator of Spatial Association (LISA) statistics were carried out on the street crime data to detect the spatial autocorrelation and clustering in the dependent variables. The spatial regression analyses were then carried out using the spatial lag model and the spatial error model on street crime rates, greenspace and sociodemographic variables.
Results: There were changing crime trends and seasonal variation of the three-street crime occurrences. Consequently, the street crime rates indicated spatial clustering with the locations of hot and cold spots for assault and robbery crime rates similar. In contrast, auto-theft crime rates emerge in different locations across the City of Toronto. Results from the spatial regression analyses show that the stem density and tree density are negatively associated with street crime rates after controlling for specific sociodemographic factors. Also, the basal area density was not significant in the spatial regression analyses on street crime rates. The six sociodemographic indicators (median household income, unemployment rate, lone parent, high school degree holders, housing units occupied by owners and renters) were significantly associated with the three street crime rates in this current study.
Conclusion: This thesis contributes to the existing literature by using a spatial-statistical approach to estimate greenspace variables and explored their relationship with street crime rates. This study draws attention to the use of specific sociodemographic factors with street crime types, and the influence parts of a tree (greenspace) could have on street crime rates across the City of Toronto. Limitations of the data were discussed, future studies concerning the recommendation of different tree species and the influence of weather on greenspace were discussed.