Comparison of the distribution maps for drug addict hotspot in Selangor using different spatial analysis tools

Abstract

The problem of drug addiction in Malaysia is worsening and causes harm to the well-being of the population of Malaysia. A report published in 2018 states that 133,684 or 0.4% of the Malaysian population are drug addicts. Furthermore, 56% of all inmates in the federal prison are locked-up because of drug-related offences. This study aim is to identify the hotspots for drug addicts in Selangor, Malaysia. This study uses three geostatistical techniques, kernel density estimation (KDE), Getis-Ord Gi*, and IDW to map the hotspots for drug addicts. The National Anti-Drug Agency (NADA) provides the data for this study which consists of 2997 cases of drug addict under supervision (DAUS) in 2016. The data are analysed using ArcGIS Pro 2.4 software. The individual DUAS represents a point vector data format with WGS 1984 Web Mercator projection. Hotspot analysis is performed using kernel density estimation (KDE), Getis-Ord Gi* and IDW.  The results show eight statistically significant hotspots for drug addicts in the sub-districts (99% confidence level and p-value < 0.001). The locations with significant hotspots for drug addicts are Bandar Serendah, Pandamaran, Bandar Klang, Bandar Kajang, Dengkil, Bandar Ampang, Bandar Damansara, and Semenyih sub-districts. This study provides spatial information that helps law enforcement agencies identify drug hotspot areas and use this information to create and enhance a defensible safe neighbourhood.  The outcome of this study facilitates law enforcement through better strategic planning for reducing drug addict hotspot areas.

Keywords: drug addicts, drug hotspots and mapping, geographical information system (GIS), Getis-ord Gi*, inverse distance weighted (IDW), kernel density estimation analysis (KDE)