Exploring Crime Statistics

Abstract

The objective of this chapter is to analyse the spatiotemporal patterns of crime in the Dhaka Metropolitan Area (DMA). Crime data for the period of August 2011 to July 2012 were acquired from Dhaka Metropolitan Police (DMP). Apart from descriptive statistics, spatial pattern analysis was carried out through local indicators of spatial autocorrelation (LISA). Linear regression was conducted using the crime incidence rate as a dependent variable and ten socio-demographic factors as potential explanatory variables. Descriptive analysis showed distinct temporal variation with the pre-monsoon season having the highest crime occurrences. Furthermore, there are three single-centred criminal activity hotspots of crime. Multivariate regression analysis showed that the size of male population and poverty rate were the best predictors of crime incidence in DMA; however, simple regression suggested that total population, number of males and size of unemployment population in the police districts were also suitable predictors. To our knowledge this is the first study that employed spatial techniques to analyse crime in DMA, and the results of this work should be valuable for informed decision-making for the law enforcing agency as well as relevant authorities whose task is to prevent increasing criminal activities in Dhaka.