MODELING URBAN CRIME PATTERNS USING SPATIAL SPACE TIME AND REGRESSION ANALYSIS

Abstract. The population size, population density and rate of urbanization are often crediting to contributing increasing a crime pattern specially in city. Urbanism model stating that the rise in urban crime and social problems is based on three population indicators namely; size, density and heterogeneity. The objective of this paper is to identify crime patterns of the hot spot urban crime location and the factors influencing the crime pattern relationship with population size, population density and rate of urbanization population. This study employed the ArcGIS Pro 2.4 tool such as Emerging Hot Spot Analysis (Space Time) to determine a crime pattern and Ordinary Least Squares (OLS) Regression to determine the factors influencing the crime patterns. By using these analyses tools, this study found that 54 (53%) out of 102 total neighbourhood locations (2011–2017 years) had a 99 percent significance confidence level where z-score exceeded +2.58 with a small p-value (p < 0.01) as the hot spot crime location. The result of data analysis using OLS regression explains that combination of exploratory variable model rate of urbanization and population size contributes 56 percent (R² = 0.559) variance in crime index rate incident [F (3,39) = 18.779, p < 0.01). While the population density (β = 0.045, t = 0.700, p > 0.10) is not a significance contributes to the change in crime index rate in Petaling and Klang district. The importance of the study is useful information for encouraging government and law enforcement agencies to promote safety and reduce risk of urban population crime areas.