Geocoding Accuracy Considerations in Determining Residency Restrictions for Sex Offenders

• NCJ Number: 226365
• Journal: Criminal Justice Policy Review Volume: 20 Issue: 1 Dated: March 2009 Pages: 62-90
• Author(s): Paul A. Zandbergen; Timothy C. Hart
• Date Published: March 2009
• Length: 29 pages
• Publication Format: Article
• Publication Type: Report (Study/Research); Legislation/Policy Analysis
• Language: English
• Country: United States of America

Annotation

This case study of sex offenders in Orange County, FL, characterizes the positional accuracy of street geocoding and its impact on spatial analysis when used to establish the residency restriction zones for sex offenders under residency restriction laws that prevent specified sex offenders from living within a certain distance of geographic areas where children regularly congregate, such as schools and day-care centers.

Abstract

The findings show that positional errors in street geocoding were substantial and may bias conclusions drawn from proximity analysis. This indicates that street geocoding is not appropriate for assessing residency restriction violations for sex offenders. Address geocoding is the process of creating an XY location of an address (associating an address record with a point on a map). In the most common approach to address geocoding, a street network is represented as street line segments that hold street names and the range of house numbers on each side of the street. Geocoding is achieved by first matching the street name, then the segment that contains the house number, finally placing a point along the segment based on linear interpolation within the range of house numbers. Collectively, the current body of knowledge suggests that the positional errors in street geocoding can be substantial and must be characterized in a manner relevant to the use of the street geocoding results. This study explains why the positional error for the schools in this case is so much higher than normal. The geocoding of sex offenders’ residence was also found to have substantial positional errors, such that the true number of offenders living within the 1,000-foot restricted zone around schools and day-care facilities was underestimated. The use of property boundaries is the preferred technique and produces reliable results. 11 figures, 4 tables, 10 notes, and 44 references