Law Enforcement and GIS

Here's a review of an article in a text book which shows the relationship between GIS and law enforcement.  The North Carolina, High Point police department was highlighted for its intervention program. Kernal Density is the primary analysis technique for the project.

Participation Article Review-  GIS & Law Enforcement

Review by Karen F. Mathews.  GIS Applications, 5100 L

Article:   Chapter 24, Using GIS to Identify Drug Markets and Reduce Drug Related Violence.   Article was presented  as a Summary of an award given for the 2006 Herman Goldstein Award for Problem Oriented Policing.    Subtitle: A Data-Driven Strategy to Implement a Focused Deterrence Model and Understand the Elements of Drug Markets.

Author:   Eleazer D. Hunt, Marty Sumner,  Thomas J. Scholten and James M. Frabutt

Law enforcement officials in High Point, North Carolina sought to break the cycle of drug related crimes and the violence connected to drug dealing in their small mountain community.  A problem found nationwide in largest of cities and the smallest towns.  The accepted modius operandi to address the drug problem nationwide had been the combined methods of surveillance, undercover drug buys and routine mass drug sweeps by law enforcement.  Instead of lowering the presence of drug activity these standard procedures over time have increased community suspicion and distrust for law enforcement officials in general.    High Point’s goal was to use a new approach to an old problem.   Breaking the cycle of police distrust, drug activity and creating a continued community ownership model of expected neighborhood behavior is the nexus of the data-driven focused deterrence method described in this article.   The objective is using an intervention deterrence program instead of a punishment model.  However, it is the essential GIS techniques which were used to develop an objective data set for the target intervention zone that made the project “doable”, successful and later replicatible.

To do an intervention, the High Point department needed to find the most suitable neighborhood zone and turned to GIS to assist with the task.  A decision was made to use a single prior year as the source of data collection.  Data consisted of  911 calls, police reports, drug arrest, field contacts and specific drug related crimes categorized as serious.  GIS changed the data use approach from a purely police perspective  “Where are the drugs?, Let’s go there;” to a workflow question: “Where are there densities of violent, sex or weapons crimes spatially concurrent with drug sales?” (pg 398).  Kernel density maps made on each layer were used as the primary analysis technique.  As described in the article, several initial hypothesis were changed by the use of GIS in selecting the intervention neighborhood:

1)  False hot spots were identified as default report locations;

2) most serious crime arrests were not associated with drug activity;  

3) the housing complex wasn’t the best place to have an intervention;  and

 4)  a smaller number of individuals were directly involved in the sale of drugs.

Selecting the exact zone required cooperative efforts between GIS analysis and selected police activity to locate the particular community and persons to be approached with the intervention techniques.  GIS was able through spatial analysis to develop a visual spatial structure of a drug market.  An intervention site was chosen in the West End neighborhood and carried out. 

When High Point, North Carolina law enforcement chose to use GIS to implement a data driven focused deterrence model for their drug related violence, they did not expect it to become a philosophy for their departments.   Further post analysis revealed a 31% decrease in drug offenses and 37 % reduction in violent crimes.  Community attitudes toward law enforcement have improved significantly through a post study of the 911 calls.  Calls increased and changed in their nature from drug related issues to quality of life calls.  Post intervention kernel density studies were conducted using the same selection