II. A Brief History of Crime Mapping
Interestingly, the earliest efforts at crime mapping can be traced to the roots of the discipline of criminology itself. In the early 19th century, a number of studies examined the distribution of crime in France and England. Brantingham and Brantingham (1991a) provided an overview of some of the findings of the main studies from this era. Guerry and Quetelet mapped crimes in France at the department level and found that crimes were not distributed evenly across departments. They also found that there was stability over time in both areas with high crime and areas with low crime over time. These findings were echoed in England with studies by Plint, Glyde, and Mayhew.
In the United States, Shaw and McKay’s (1942) seminal study of juvenile delinquency in Chicago made extensive use of crime maps. Shaw and McKay borrowed Park and Burgess’s (1924) ecological model and divided the city into five different zones. They found that the zone adjacent to the central business district, the zone of transition, perpetually suffered from the highest rates of juvenile delinquency and other social problems regardless of the specific ethnic group occupying the zone at the time. This research was instrumental in popularizing social disorganization theory and inspired a number of similar mapping projects in Chicago; Philadelphia; Richmond, Virginia; Cleveland, Ohio; Birmingham, Alabama; Denver, Colorado; Seattle, Washington; and other cities.
Accompanying these early efforts in crime mapping were developments in the profession of policing that provided additional opportunities for crime mapping. In the late 19th and early 20th centuries, the professionalization movement in policing encouraged police organizations to compile statistics documenting the extent of crime in their jurisdictions. In fact, one of the main justifications for the creation of Federal Bureau of Investigation was for the explicit purpose of documenting the extent of crime in the United States through the Uniform Crime Reporting program (Mosher, Miethe, & Phillips, 2002). During this time, many agencies began compiling crime statistics and conducting analyses of crime data. Crime mapping was primarily done using pin maps, which were very time-consuming and provided only a basic visualization of crime patterns.
The late 1960s and early 1970s were critical for the development of crime mapping. In 1966, the Harvard Lab for Computer Graphics and Spatial Analysis developed SYMAP (Synagraphic Mapping System), one of the first widely distributed computerized mapping software programs. The Environmental Science and Research Institute was founded in 1969 and in the subsequent decades emerged as one of the top distributors of GIS software, including the current ArcView and ArcGIS software packages. Also around this time, the U.S. Census Bureau began the ambitious GBF-DIME (Geographic Base Files and Dual Independent Map Encoding) project, which was used to create digitized street maps for all cities in the United States during the 1970 census (Mark, Chrisman, Frank, McHaffie, & Pickles, 1997). These advances were necessary for the development of GIS programs used in crime mapping.
The use of GIS programs for mapping has been the most important advance in the field of crime mapping. There are several important advantages in using virtual maps instead of physical maps. First, computers have dramatically reduced the time and effort required to produce crime maps. Given the relatively low cost and user-friendliness of many of these software programs, it no longer requires a substantial investment for agencies that wish to engage in crime mapping. Second, these GIS programs reduce the amount of error associated with assigning geographic coordinates to crime events. Third, virtual maps are much more flexible than physical maps, allowing researchers and crime analysts to compare the geographic distribution of crimes against other characteristics of the area under investigation (e.g., census bureau information, city planning and zoning maps, and maps produced by other agencies). Finally, GIS and other spatial analysis software provide powerful statistical tools for analyzing and detecting patterns of criminal activity that cannot be detected through simple visual inspection.
In the mid-1970s and early 1980s, a crisis of confidence in traditional police practices emerged following the results of studies, such as the Kansas City Preventative Patrol experiment, that suggested that the police were not effective in combating crime (Weisburd & Lum, 2005). Goldstein’s (1979) problem-oriented policing emerged as a response to this crisis and emphasized that policing should involve identifying emerging crime and disorder problems and working to address the underlying causes of these problems. Academic interests in the field of criminology also began to shift during this time. While many criminologists were concerned with causes of crime that were outside the sphere of influence of police agencies (e.g., economic depravation, differential association, and social bonds), a number of researchers, such as Jeffery (1971), Newman (1972), and Cohen and Felson (1979), began discussing factors that contribute to the occurrence of crime that were more amenable to intervention. The combination of the shift in theoretical focus in criminology and the shift in the philosophy of policing yielded new opportunities for crime mapping and initiated a resurgence of research on both the geography of crime as well as crime prevention strategies involving crime mapping.
Although the first instances of computerized crime mapping occurred in the mid-1960s in St. Louis, Missouri, the adoption of computerized crime mapping across the United States remained relatively slow. Although a number of agencies, in particular in larger jurisdictions, became early adopters of computerized crime mapping technology, the large period of growth in computerized crime mapping did not begin until the late 1980s and early 1990s (Weisburd & Lum, 2005). The rate of adoption of crime mapping among departments greatly increased as desktop computers became cheaper and more powerful and GIS software became easier to use and more powerful. The Compstat program, which started in 1994 in New York City, emphasized crime mapping as a central component to strategic police planning and helped popularize crime mapping among police agencies. With assistance from the Office of Community Oriented Police Services and the National Institute of Justice, a large number of departments adopted computerized crime mapping practices. By 1997, approximately 35% of departments with more than 100 officers reported using crime mapping (Weisburd & Lum, 2005).