Reference data and geocoding quality: Examining completeness and positional accuracy of street geocoded crime incidents

Abstract

urpose - The purpose of this paper is to examine the influence of reference data, input address quality, and crime type on completeness and positional accuracy of street geocoded crime events. Design/methodology/approach - Existing data were analyzed using ArcGIS, including crime incident information, street network reference data, and address point and/or parcel reference data. Geocoding completeness was determined by the overall match rate. Positional accuracy was determined by comparing the Euclidian distance between street geocoded locations of crime events to the corresponding address point/parcel geocoded location. Findings - Results indicate that match rates vary by reference data, input address quality, and crime type. Local street centerline files consistently produced match rates that were as good as - and in many cases superior to - other types of reference data, including commercial data. Greater variability in positional accuracy was observed across reference data when crime type and input address quality was considered, but results were consistent with positional accuracy analysis conducted using data from other disciplines. Practical implications - Results provide researchers and practitioners with valuable guidance and insight into one of the most basic - albeit fundamental - procedures related to the spatio-temporal analysis of crime, suggesting that reference data required to produce geocoded crime incidents successfully and of high quality does not necessarily mean a large financial investment on the part of law enforcement agencies or researchers interested in the geospatial analysis of crime. Originality/value - Prior to this investigation, a comprehensive examination of the impact of data quality on geocoded crime events was absent from the literature.