+1 (208) 254-6996 essayswallet@gmail.com
  

Imagine that you are a crime analyst. You will be creating a geographic crime map that spans three years of crime data.

You  will create a CSV file, using an excel document, saved as a CSV file.  To build the CSV file you will need to do the following:

  1. Go to the LexisNexis Community Crime Map site: http://communitycrimemap.com/
  2. Choose a crime you would like to search. (e.g., theft, robbery, arson).
  3. Select  a geographic area that has 5-10 instances of the specific crime you  selected, which occurred in each of the 3 years preceding last year  (e.g., if you are taking this course in 2018, then you would search for  the specific crime in 2014, 2015, and 2016). NOTE: you want to use a  geographic area with a small data set, e.g., a city of a million or more people would likely have a large data set.
  4. Identify  5-10 instances of the crime for each of the 3 years. Click on the icon  of the crime incident to ascertain the crime data. For a tutorial on  using the crime map, watch the video from the Madison Police department  explaining its use: https://www.youtube.com/watch?v=2c09YubbaD8&feature=youtu.be
  5. Using  the excel spreadsheet that you will save as a CSV file, create the  following columns and input your crime data you have gathered from the  crime map:
  • IR number (“ID” in CSV file)
  • Name of Crime (“class” in CSV file)
  • The  date and time the crime occurred (“Events time” in CSV file). You must  enter the date and time in the following format: yyyy-mm-ddThh:mm:ss
  • Latitude of crime location (“Point y” in CSV file)
  • Longitude of crime location (“Point x” in the CSV file)

NOTE: Access the “How to find Latitude and Longitude” document in the Topic 5 Topic Materials as a resource for this assignment.

Conduct  analysis on the geographic data using the techniques from the assigned  readings. Write a 750-1,000-word summary on the analysis of the data.

Be sure to cite three to five relevant scholarly sources in support of your content.

Utilize the Hunchlab website as a resource for the Topic 5 assignments.

URL:https://www.hunchlab.com/ 

Utilize the ArcGIS website as a resource for the Topic 5 assignments.

URL:http://www.esri.com/arcgis/about-arcgis 

Utilize the Maptitude website as a resource for the Topic 5 assignments.

URL:https://www.caliper.com/maptitude/crime/default.htm

JUS-636 Topic 5 How to Find Latitude and Longitude

In order to locate the Latitude and Longitude of an address:

1. Put the address into google (maps). Note, if the address contains “x” use “0” as a substitute. For example, if the address is 33xx Camelback road, type in 3300 Camelback road, click “Search”

2. When the address appears (it will be a red blob that looks like a rain-drop), right-click on it. Click on “What’s here?”

3. A small box will appear (see screen shot below)

4. The numbers below the address are the Latitude (33.512802) and Longitude

(-112.127409). Latitude is always listed first, and Longitude second.

March 15, 2013 Page 1 of 3

How to Attach Multiple Documents to an Assignment Submission

1. Access the assignment dropbox. Click the “New Attempt” button in the lower right corner.

2. A “New Attempt” dialogue box will display. Enter the title of the attachment in the “Title” field. Click “Choose File”

March 15, 2013 Page 2 of 3

3. A file explorer window will display. Find the location of the file to be uploaded. Click on the file. Click “Open” on the file explorer window.

4. The file will display in the box below the words “Attach Documents.” To attach another file, click, “Choose File.”

5. A file explorer window will display. Find the location of the file to be uploaded. Click on the file. Click “Open” on the file explorer window.

March 15, 2013 Page 3 of 3

6. The additional file will display in the box below the words “Attach Documents.” Repeat steps 3 and 4 to attach additional files. Once all files display in the box below the words

“Attach Documents,” click “Save.”

7. Multiple files will display under the “Attached Documents” area of the assignment dropbox. To submit the files to TII, click the “Submit” under the Turnitin Report column. Once the TII

report similarity percentage displays, then click the “Submit” button under the “Final

Submission” column.

https://search-credoreference-com.lopes.idm.oclc.org/content/entry/wileycacj/crime_mapping/0?institutionId=5865

(the above is the link if you needed it)

Crime Mapping

from  The Encyclopedia of Criminology and Criminal Justice

Mapping crime has become a routine function of police departments throughout the world. All segments of the criminal justice system use aspects of crime mapping in their daily routine. In addition, crime mapping has been used to advance criminological theory.

· crime

· ecology

· environmental criminology

· law enforcement

· spatial analysis

Present day crime mapping is the application of a geographic information system (GIS) in order to understand crime data and the utilization of spatial analysis of crime situations in criminal justice agencies, especially in police work. Even though computer technology has been developed within the last century, the early crime mapping process goes back to the eighteenth century (Boba 2005). In one of the first instances of crime mapping, crimes against property and persons were plotted on a map in reference to the level of education of the offenders in France. The process of mapping crime was later used in England to study crime situations using area characteristics, such as urbanization and highway proximity (Paulsen and Robinson 2004).

The practice of crime mapping was initiated in the United States in the early 1920s by the Chicago School. It was done for the purpose of studying the factors behind juvenile delinquency in Chicago. All crime mapping in these early years was done by hand and continued to be done by hand for many decades. It was not until the early 1960s that computers were used to help with the process. In the 1980s the advent of client server technology made GIS more accessible to the police and research communities. For instance, the National Institute of Justice teamed up with researchers and practitioners in five cities (Jersey City, Harfort, San Diego, Pittsburgh, and Kansas City) to use innovative tools in studying drug markets and tracking their movements over time.

Since the 1990s, with the advancement of computer technology as well as police data systems, crime mapping has become a more accessible and practical tool for police practitioners and researchers. GIS software can be installed on desktop computers that have sufficient power to process a large batch of data in seconds. It was during this period that data on crimes, arrests, accidents, and calls for service became available in an electronic format, through either electronic dispatch systems or records management systems. Recent developments in digital mapping (including affordability) and technological advancements in policing have enabled environmental criminologists to explore the special dimensions of criminality in comprehensive ways using crime mapping.

The crime mapping process requires elements of a computer system not much different from other computerized processes. These are computer hardware, software, data, peopleware, and work procedures involving organization policies, safety protocols for the system, and so on. However, there are some requirements specific to computerized crime mapping. For example, certain devices may be needed in the working process (such as a digitizer, a plotter, and a GPS—global positioning system—device). People working in crime mapping should be equipped with backgrounds in GIS technology or have access to a GIS specialist, if needed.

Software products for crime mapping are quite numerous. Some have been created specifically for law enforcement uses, including Crime Stat and RAIDS. Other programs are geographic information software that can be used for the general purpose of mapping. Like other types of software, software for crime mapping can be divided into two categories: open source and proprietary. Open source mapping software is available to any user for little or no fee, and includes such programs as GRASS GIS and SAGA GIS. Proprietary mapping software must be purchased from the provider and includes ArcGIS, Map 3D, and Bentley Map, among others.

Creating a crime map incorporates working with data layers. Data involved in the mapping process consist of three parts. The first part is the structure of the layout which is the base map. It contains the basic information (points, lines, and polygons) that make up a map, just like a common map. This base map can be used for mapping in police work, as well as for other purposes of mapping, depending on the user’s needs. The second part, crime data, includes the location and details of the crime incidents that the users intend to incorporate into the crime map. These details may include incident type, date and time of the incident, address or location of the incident, property loss, and so on. This information can be stored in and retrieved from a separate data server or at the same location as the other data used to create the map. The last part is any other data which the users decide to show in the crime map, including points of interest, links to other media (e.g., video clip, audio file, and picture), and other information useful for presenting the map so that it can more easily be understood by viewers.

The process of making a crime map involves putting the data layers together. Recent advances in computer technology, relevant software, and digitized data make it possible to create a crime map digitally. One process crime map creators often need to operate is geocoding, which is the process of transforming the locations of crime data into geographic coordinates, usually in the form of latitude and longitude, ready for crime mapping software to utilize. A geocoder, which could be a piece of software or a web service, can do the transformation for the researcher. Crime mapping always involves a large amount of crime and geographic data (more than 10,000 data points are not unusual), making a geocoder a very useful tool. The process of transforming crime locations into coordinates is called address interpolation.

Many theories of crime try to explain individual or group criminal behavior and ignore other contributors to crime, including the environment. Theorists who want to include environmental factors to explain crime have begun to see crime as an event—not just a product of an interaction between people, but an interaction between victims, offenders, and their environments (Paynich and Hill (2009): 97–118). Paynich and Hill mention that after researchers began to examine the contributions of time and space to criminal events, crime mapping became an important tool for both crime fighters and criminologists.

Ecological theories look for explanations of individual actions in general features of the social structure in which an individual is embedded. The social ecology perspective evolved from the early social ecologists in France in the mid 1800s, to the Chicago school in the 1920s, and finally to a recent revival in contemporary studies in the ecology of crime. The social ecology perspective developed into more specifically focused, place-based theories of crime, particularly routine activities theory.

Bernard Cohen did a study (1980) on street-level prostitution and identified “hot spots” of prostitution activity. Cohen’s work is one of the first empirical studies to document the spatial and temporal intersection of “motivated” offenders and the crime-facilitating properties of place proposed by the routine activities theory theorized by Cohen and Felson (1979). Modern GIS capabilities, combined with point data on the locations of individual crimes, make studies like those described here feasible by enabling researchers to routinely obtain measures of crime variables at these nontraditional and smaller levels of aggregation. Consequently, several different theoretical approaches have been made possible, including place-based theories derived from routine activities theory (Cohen and Felson 1979) and rational choice theory (Cornish and Clark 1986), crime pattern theory (Brantingham and Brantingham 1993), and Crime Prevention Through Environmental Design (CPTED).

Mapping crime has become a routine function of police departments throughout the world. Moreover, all segments of the criminal justice system, from courthouses to correctional institutions, from the probation officer’s desktop to the operation commander’s handheld, uses a variety of crime maps. Hot spot to offender behavior, susceptibility analysis to density analysis, and other crime mapping procedures are part of their daily routine (Can and Leipnik 2010). Similarly to other general types of data, crime maps can be used by law enforcement to present crime information in many ways. One way to classify the maps used by law enforcement is to look at it from a crime analysis standpoint. Crime analysis aligns mapping used by law enforcement into three basic areas: administrative, strategic, and tactical (Paynich and Hill 2009). Under these areas, six categories of mapping types are commonly used: single-symbol mapping, buffers, graduated mapping, chart mapping, density mapping, and interactive crime mapping. Single-symbol mapping, as suggested by its name, utilizes one type of symbol to represent features (e.g., locations of police stations, schools, incidents, and so on). This type of mapping is not suitable for data with multiple incidents occurring at the same specific locations. A buffer map displays a specific area surrounding a feature, making it possible to find other features located in that area. It is suitable for analysis purposes. For instance, an analyst can use a buffer map to show the number of drug-related incidents occurring within a distance of 1,000 feet from schools in a county during a six month period. Graduated mapping utilizes different sizes or colors of features to represent particular values of variables. Differences in size or color reflect variation in value. Chart mapping incorporates charts (pie and bar) into the map. Density mapping conveys the frequency of incidents in specific areas through color shading, with darker colors representing more incidents. Interactive crime mapping is usually offered to users via internet by police departments. The interactive functions enable users, police and citizens, to conduct some basic crime mapping independently. (See Figures (1234).)

Source: Salih Can and Prapon Sahapattana.

Locations of murder incidents in police precinct 67, Brooklyn. The map indicates that there were 50 murder incidents in the precinct in recent years. Of these incidents, 26 (52%) occurred at avenues.

Source: Salih Can and Prapon Sahapattana.

Weapon distribution of murder incidents in police precinct 67, Brooklyn. Of the 50 incidents, 46 (92%) were committed with a handgun, two with a blunt instrument, one with a knife, and one by fire.

Source: Salih Can and Prapon Sahapattana.

Victim gender distribution of murder incidents in police precinct 67, Brooklyn. Victims were male in 44 (88%) incidents and female in 6 (12%) incidents.

Source: Salih Can and Prapon Sahapattana.

Locations of murder incidents in police precinct 67, Brooklyn. Outside buildings, 29 (58%); inside buildings, 19 (38%); no data, 2 (4%).

Two approaches to analyzing crime situations are using exact locations of incidents to generate hot spots of crime and using tabular data and statistics to show the analysis. “Hot spots” are concentrations or clusters of crimes in space. There are three fundamental ways of identifying crime hot spots; visual inspection, statistical identification, and the theoretical approach. Crime can be mapped in areas or layers (for instance, by socioeconomic status, traffic intersection, or location of places) and then put on top of each other. Viewing these layers together makes it possible to see the natural patterns of movement for the people who use the area, the location of potential offenders, and likewise potential targets. This provides a reasonable initial explanation for a concentration or clustering of crimes. Brantingham and Brantingham (1999) find it helpful to describe these levels or layers of crime potential as the environmental setting, with a more detailed description of normal movement patterns, location of potential crime generators, location of potential crime attractors, characteristics of place, and ecological labels. Hot spots are also directly related to structural setting, which is composed of relatively slowly changing elements (e.g., laws, social norms, class structures, the division of labor, the physical layout of the built environment, the local topography, the local climate, and so forth) and the activity setting, which is composed of routines and adventures. The activity setting is shaped by mobility and awareness factors that are tied to personality, age, social status, income, education, and a variety of other social and psychological characteristics, as well as to positioning in physical space-time.

For purposes of analysis, a statistical classification is beneficial for crime analysts both to understand the crime situation and to present the situation to the audience. That is, the data in focus will be categorized into groups for investigation. A statistical classification is used in crime mapping in order to determine how the data on the map will be displayed in the way that is helpful to the viewers to more clearly understand the data. Generally, four statistical classifications can be found in the crime mapping process: natural breaks, equal interval, quartile, and standard deviation. The type of classification is commonly exhibited in the legend of the map.

Natural breaks classification is the default in many geographic information system programs and also the most popular in crime analysis. Generally, it is exercised to explore the data first and also to obtain a quick analysis of the data. Equal interval classification could be used by a crime analyst for data with the same number of incidents in each interval. The nature of crime incidents, however, does not follow this pattern. Consequently, this type of classification is not often used for the final crime map. A quartile map is used for comparison between data sets in different maps. In each map, the different symbols are displayed in order for viewers to see the crime situation on the map. Finally, a map using the standard deviation classification involves utilizing the means and standard deviations of the data to establish the break points of the groups. This type of map is useful for comparison purposes between different data sets. Finally, an analyst could use a manual method to decide the break points. The method utilized to create a map and how many categories to use, should be determined by the analyst, according to the purpose of and audience for the map.

Crime mapping is unquestionably salient and accepted among law enforcement agencies. Mapping crime data, as well as data related to disorder problems, enables police officers to learn the location of crime or disorder incidents. It also reveals the reasons behind the problems through the process of crime analysis. Research shows that a great number of large police departments have been using computerized crime mapping for almost 15 years (Mamalian & LaVigne 1999). Research also confirms the use of computerized crime mapping by small departments, as well. They use crime maps for staff allocation decisions, evaluation of intervention programs, informing citizens about crime situations in their neighborhoods, and identification of repeat crime locations, as well as for calls for service.

The rapidly decreasing price and user friendly nature of desktop computers and GIS software has made crime mapping much more prevalent. Currently, the integration of crime mapping with GPS has increased the accuracy of the incident locations. People are increasingly familiar with the integration of crime mapping due to web-based applications and multimedia, such as audio, video, and pictures. Adding advanced statistical programs to the picture, more sophisticated analyses are rendered more advantageous to officers and citizens. The trend of crime mapping, as well as crime analysis, is likely to continue to grow in law enforcement agencies, especially as new technologies to enhance the utility of this practice are developed.

SEE ALSO: Human EcologyRational Choice TheoryRoutine Activities and CrimeSocial Disorganization Theory.

References

 Boba, R. (2005) Crime Analysis and Crime Mapping. Sage Thousand Oaks CA. 

 Brantingham, P.; Brantingham, P. (1993) Nodes, path and edges: Considerations on the complexity of crime and the physical environment. Journal of Environmental Psychology 13, 3-27. 

 Brantingham, P.; Brantingham, P. (1999) A theoretical model of crime hot spot generation. Studies on Crime and Crime Prevention 8(1), 7-25. 

 Can, S. H.; Leipnik, M. R. (2010) Use of geographic information systems in counterterrorism. Professional Issues in Criminal Justice 5(1), 39-54. 

 Cohen, B. (1980) Deviant Street Networks: Prostitution in New York City. D. C. Heath and Co Lexington MA. 

 Cohen, L.; Felson, M. (1979) Social change in crime trends: A routine activity approach. American Sociological Review 44, 588-608. 

 Cornish, D.; Clark, R. (1986) The Reasoning Criminal. Springer New York. 

 Mamalian, C. A.; LaVigne, N. G. (1999) The Use of Computerized Crime Mapping by Law Enforcement: Survey Results. US Department of Justice. https://www.ncjrs.gov/pdffiles1/fs000237.pdf, accessed January 29, 2013. Check for full text

 Paulsen, D. J.; Robinson, M. B. (2004) Spatial Aspects of Crime: Theory and Practice. Allyn and Bacon Boston. 

 Paynich, R.; Hill, B. (2009) Fundamentals of Crime Mapping. Jones & Bartlett Sudbury MA. Check for full text

Further Readings

 Bowers, K.; Hirschfield, A. (2001) Mapping and Analyzing Crime Data. Taylor & Francis London. 

 Canter, D. (2003) Mapping Murder: The Secrets of Geographical Profiling. Virgin Books London. 

 Chainey, S.; Ratcliffe, J. (2005) GIS and Crime Mapping. John Wiley & Sons, Inc Hoboken NJ. 

 Chainey, S. P.; Thompson, L. (2012) Engagement, empowerment and transparency: Publishing crime statistics using online crime mapping. Policing 6(3), 228-239. 

 Clarke, R. V.; Eck, J. (2003) Become a Problem Solving Crime Analyst. University College London, Jill Dando Institute of Crime Science London. 

 Davis, B. E. (2001) GIS: A Visual Approach. OnWord Press Albany NY. 

 Harries, K. (1999) Mapping Crime: Principle and Practice. US Department of Justice Washington, DC. 

 LaVigne, N.; Wartell, J. (2000) Crime Mapping Case Studies: Successes in the Field, vol. 2. Police Executive Research Forum Washington, DC. 

 Leipnik, M. R.; Albert, D. P. (2003) GIS in Law Enforcement: Implementation, Issues and Case Studies. Taylor & Francis London. 

 Levine, N. (2004) Crimestat III: A Spatial Statistics Program for the Analysis of Crime Incident Locations. Ned Levine and Associates Houston TX. 

 Rich, T.; Shively, M. (2004) A Methodology for Evaluating Geographic Profiling Software: A Report Prepared for U.S. NIJ. Abt Associates Cambridge MA. 

 Rossmo, K. (2000) Geographic Profiling. CRS Press Boca Raton FL. Check for full text

Salih Hakan Can Prapon Sahapattana

Wiley Copyright © 2014 by John Wiley & Sons, Ltd.

· APA

· Chicago

· Harvard

· MLA

Can, S. H., & Sahapattana, P. (2014). Crime mapping. In J. S. Albanese, Wiley series of encyclopedias in criminology and criminal justice: The encyclopedia of criminology and criminal justice. Wiley. Credo Reference: https://lopes.idm.oclc.org/login?url=https://search.credoreference.com/content/entry/wileycacj/crime_mapping/0?institutionId=5865

The Encyclopedia of Criminology and Criminal Justice

Previous Article Next Article

This encyclopedia offers a state-of-the-art survey informed by the very latest theory and research. It combines this breadth of coverage with the authority and international perspective of an experienced editorial team, creating a definitive reference resource for students, scholars, and professionals in these interdisciplinary and dynamic fields.

Author(s): Jay S. Albanese

Edition: 1st

Articles: 550

Images: 9

People: 604

Related Searches

Geographic information system Crime mapping Crime analysis Criminology Crime Law enforcement techniques Police Public participation geographic information system Crime data Criminal justice

Related Articles

Full text Article

Crime Mapping

in  World of Criminal Justice, Gale

The term, computerized crime mapping, refers to technology that enables law enforcement agencies to analyze and correlate crime data within a…

lable at ScienceDirect

Applied Geography 79 (2017) 203e211

Contents lists avai

Applied Geography

journal homepage: www.elsevier .com/locate/apgeog

The crime kaleidoscope: A cross-jurisdictional analysis of place features and crime in three urban environments

Jeremy D. Barnum a, *, Joel M. Caplan a, Leslie W. Kennedy a, Eric L. Piza b

a Rutgers Center on Public Security, Rutgers University School of Criminal Justice, Newark, NJ, USA b John Jay College of Criminal Justice, City University of New York, USA

a r t i c l e i n f o

Article history: Received 25 May 2016 Received in revised form 17 October 2016 Accepted 22 December 2016 Available online 17 January 2017

Keywords: Crime pattern theory Risk terrain modeling Spatial influence Robbery

* Corresponding author. School of Criminal Justice, 123 Washington Street, Newark, NJ, 07102, USA.

E-mail address: jeremy.barnum@rutgers.edu (J.D.

http://dx.doi.org/10.1016/j.apgeog.2016.12.011 0143-6228/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Research identifies various place features (e.g., bars, schools, public transportation stops) that generate or attract crime. What is less clear is how the spatial influence of these place features compares across relatively similar environments, even for the same crime. In this study, risk terrain modeling (RTM), a geospatial crime forecasting and diagnostic tool, is utilized to identify place features that increase the risk of robbery and their particular spatial influence in Chicago, Illinois; Newark, New Jersey; and Kansas City, Missouri. The results show that the risk factors for robbery are similar between environments, but not necessarily identical. Further, some factors were riskier for robbery and affected their surrounding landscape in different ways that others. Consistent with crime pattern theory, the results suggest that the broader organization of the environmental backcloth affects how constituent place features relate to and influence crime. Implications are discussed with regard to research and practice.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Crime can happen anywhere, but some places are more likely to experience crime than others. Research has found that a majority of police demand originates from just a few places (e.g., Braga, Hureau, & Papachristos, 2011; Braga, Papachristos, & Hureau, 2010; Sherman, Gartin, & Buerger, 1989; Weisburd, Groff, & Yang, 2012). Places are “very small micro units of analysis,” including specific addresses, groups of addresses, block faces, or street seg- ments (Weisburd, 2008, p. 2). Crime concentrates at certain places because of their unique social and physical qualities, which creates context that invites and sustains legally problematic behavior (Eck & Weisburd, 1995; Kennedy, Caplan, & Piza, 2011).

Environmental criminological theories (Wortley & Mazerolle, 2008) frame crime events within the context of the environ- mental backcloth (Brantingham & Brantingham, 1993). Distributed through this backcloth are place features, such as bars, schools, or public transportation stops that generate and attract crime (e.g., Brantingham & Brantingham, 1995). However, environments are highly complex, and thoughmany of the same features exist within different environments, their overall form and function is distinct

Rutgers University e Newark,

Barnum).

(Lynch, 1960). Kennedy (1983) refers to the kaleidoscopic organi- zation of place features about the urban landscape resulting from variety of forces (i.e., historical, cultural political, and economic) that influences its past and ongoing development. Poon (2015) posits that environments have their own “spatial DNA.” Given the relative organization of each jurisdiction’s environmental back- cloth, the spatial influence of constituent place features on crime may not necessarily generalize across environments, even for the similar types of crime.

This study compares the criminogenic spatial influence of place features in different urban environments. It is hypothesized that place features commonly assumed to correlate with crime may not have a static influence, even across similar types environments for similar types of crime. Risk terrain modeling (RTM), a geospatial crime forecasting and diagnostic tool (Caplan, Kennedy, & Miller, 2011), is utilized to identify place features that increase the risk of robbery and their particular spatial influence in Chicago, Illinois; Newark, New Jersey; and Kansas City, Missouri. The results show that the significant risk factors for robbery were similar across environments, but not necessarily identical. In other words, just because a given place feature aggravated robbery in one jurisdic- tion does not necessarily mean it did so in another. Further, some factors were riskier for robbery and affected their surrounding landscape in different ways than others. Consistent with theories of environmental criminology, the results suggest that the broadermailto:jeremy.barnum@rutgers.eduhttp://crossmark.crossref.org/dialog/?doi=10.1016/j.apgeog.2016.12.011&domain=pdfwww.sciencedirect.com/science/journal/01436228http://www.elsevier.com/locate/apgeoghttp://dx.doi.org/10.1016/j.apgeog.2016.12.011http://dx.doi.org/10.1016/j.apgeog.2016.12.011http://dx.doi.org/10.1016/j.apgeog.2016.12.011

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211204

organization of the environmental backcloth affects how constit- uent place features relate to and influence crime, which has im- plications for theories of crime and place and policy implications pertaining to the ways in which police should respond to prob- lematic places throughout their jurisdiction to achieve crime prevention.

2. Conceptual framework

In the mid-20th century Shaw and McKay (1942) observed that juvenile delinquency was unevenly distributed throughout the environmental landscape of Chicago. Specifically, they demon- strated that delinquency was highly concentrated in areas sur- rounding the center of the city and gradually declined in areas moving radially outward towards the edges of the city in a fashion consistent with the concentric zone model develop by Burgess (1925). Further, they found that delinquency remained highly concentrated in these particular areas over time, regardless of the people who lived there.5They attributed their findings to social disorganization caused by broader structural forces of communities such as poverty, residential mobility, and demographic heterogeneity.

Subsequent research had difficulty generalizing these particular spatial patterns of crime to other cities (Bursik, 1988). Following World War II there was a large-scale population movement; “Dirty industries left [the inner city] to be in the suburbs, or even other developing countries. Downtown living became a luxury, and former working class neighborhoods [were] invaded by pro- fessionals in the process of gentrification” (Andresen, 2014, p. 21). Ecological change was the norm and the concentric zone model used by Shaw and McKay to illustrate the distribution of crime in Chicago did not necessarily “fit” other cities. Indeed, alternative ecological models such as the sector model (Hoyt, 1939) and the multiple nuclei model (Harris & Ullman, 1945) were developed to describe the ecological structure of other cities. Another limitation of this research was that “environments” from this perspective largely referred to community structural characteristics of an area, rather than the physical qualities of places (Kennedy, 1983). Yet, the built environment plays an important role in organizing human behavior and thereby providing ample opportunity for crime.

Crime pattern theory (Brantingham & Brantingham, 2008) in- tegrates notions of rational choice (Clarke & Cornish, 1985) and routine activities (Cohen & Felson, 1979) to describe this relation- ship. Essentially, crime is the product of decisions about offending and the distribution of offenders, targets, and guardians, each of which are shaped by the physical environmental landscape. Spe- cifically, willing offenders are cued as they encounter viable op- portunities for crime. Decision templates provide offenders with a mechanism for recognizing and discerning good from bad targets. Decisions that lead to successfully carrying out a criminal act reinforce the template; if unsuccessful, the template is revised to avoid such decisions in the future.

Crime opportunities arise within the context of the environ- mental backcloth (Brantingham & Brantingham, 1993), which in- cludes individuals’ routine activities and the underlying networks of roads, buildings, and other infrastructure. Offenders and victims traverse the environment, engaging in their normal routines and traveling among their regular activity spaces. Crime occurs when offenders’ encounter a target that fits their decision template. Such encounters are more likely to occur at places that facilitate the

5 “Areas” is utilized intentionally. The bulk of urban ecological research has focused on large areal units, such as census tracts or blocks. In contrast, more recent research has focused specifically on “places” as described by Weisburd (2008).

“overlapping lifestyles or spatio-temporal movement patterns” of offenders and targets (Brantingham & Brantingham, 2008, p.87). Certain places do somore than others because they contain features that generate or attract crime (Brantingham & Brantingham, 1995). Crime generators concentrate a large number of people, both po- tential offenders and victims, in specific locations at the same time. Crime is more likely at crime generators because of the large number of interactions that take place. Conversely, crime attractors specifically draw motivated offenders given well-known criminal opportunities.

Place features that generate and attract crime are distributed throughout the landscape along various paths, or the routes people take (e.g., roads, sidewalks, etc.) and edges, or distinct changes in the landscape (e.g., railways, changes in land use, neighborhood boundaries, etc.), which create nodes, or areas of intense activity (Brantingham & Brantingham, 1984). The distribution of these features throughout each jurisdiction’s environmental backcloth is unique as the result of various processes involving local policies and regulations with regard to zoning, infrastructure, and urban plan- ning. Physical landscapes are constructed around natural terrains and molded around particular social, cultural, historical, and eco- nomic systems all of which influence their unique form and func- tion and ongoing change and development. The combination of these forces ensures distinctiveness in the image of cities and the ways in which behavior within them unfolds (Lynch, 1960). Kennedy (1983, p.11) conceptualizes this through the analogy of a kaleidoscope (see Fig. 1). The kaleidoscope represents an environ- ment (e.g., City A) and the shards of glass embody place features (e.g., bars, restaurants, public transportation stops) within that environment. The arrangement of place features encompasses an environment’s form. Moving from one environment to the next (e.g., from City A to City B), or turning the kaleidoscope, alters the form of that environment. Central to the analogy is that the patterning of features varies between environments, but the parts and processes that create these patterns are the same. Thus, it is the particular combinations of features at places in different environ- ments that must be identified to understand the distribution of behaviors and crime.

In sum, early ecological research demonstrated that crime is more likely in some areas of a city compared to others and sug- gested that there is value in considering what it is about those areas, beyond the individuals that exist there, that foster illegal behavior. However, this perspective primarily focused on commu- nity structural characteristics and largely neglected the influence of the physical features of environments on crime. Modern advances in data and technology have allowed researchers to demonstrate that crime is highly concentrated at very specific places throughout the geographic landscape. In this regard, several perspectives have emerged, falling under the broader realm of environmental crimi- nology to provide a theoretical basis to this phenomenon. These perspectives discuss how physical place features throughout the environmental backcloth can generate or attract crime by struc- turing the everyday routines of individuals and creating good op- portunities for offending. However, each jurisdiction has a unique backcloth and the particular ways in which certain features come together to create conditions for illegal behavior may not gener- alize, even for the same crime. Therefore, it is important to identify these patterns within the environmental backcloth of each juris- diction to better understand the more localized spatial dynamics of crime.

3. The study

The purpose of this study is to examine the physical landscapes of different environments and their relative influence on crime.

Fig. 1. (A) The kaleidoscope (i.e., environment) is composed of shards of glass (i.e., place features). The image (i.e., environmental form) that emerges is the result of the particular arrangement of glass shards. Turning the kaleidoscope (i.e., viewing another environment) alters this image. Crime concentrates at certain places within the environment, given the confluence of certain features that come together and create conditions that are conducive to offending.

6 Funded by the National Institute of Justice (Award #2012-IJ-CX-0038). 7 http://www.infogroup.com/.

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211 205

More specifically, we utilize RTM to diagnose and compare crimi- nogenic place features and their spatial influence on robbery in Chicago, Newark, and Kansas City. Consistent with environmental theories of crime it is assumed that certain place features will in- crease the risk for robbery. However, it is hypothesized that these features and their influence will vary across environments because they are uniquely patterned throughout each jurisdiction’s envi- ronmental backcloth. Fig. 1 illustrates how place features are distributed throughout the environmental backcloth in a way that is conceptually similar to the shards of glass in a kaleidoscope. Crime concentrates at places within any given environment given the confluence of features that create conditions that are conducive to offending. However, observing another environment, or turning the kaleidoscope, alters these patterns and the resulting image. As environments change, so too do the distributions of people, of- fenders, targets, and guardians, and their routines, and the ideal opportunities for crime. Thus, it is important to examine the unique ways in which the environmental backcloth of different jurisdic- tions allows crime to emerge and persist.

3.1. Risk terrain modeling

The premise of RTM is that certain places have particular fea- tures that, when combined in prescribed ways, create context in which crime becomes more likely (Caplan et al., 2011, p. 365). The purpose of RTM is to diagnose place-based risk factors for crime and then identify where they are collocated to increase vulnera- bility to crime (Kennedy, Caplan, Piza, & Buccine-Schraeder, 2015). The process of RTM involves a general series of steps beginning with the selection of an outcome event, a study setting, and time period. Then, all potentially risky place features are identified. The spatial influence of each feature is operationalized to a continuous surface of raster GRID cells in a geographic information system

(GIS). This produces a set of separate layers representing the spatial influence of each feature at every micro-level place (i.e., cell). Each risk layer is then empirically tested, weighted, and then combined with other statistically significant layers to create a composite risk terrain layer with each cell containing a value indicating the spatial influence of all risky place features throughout the entire geography.

3.2. Data and study settings

All data for this study were obtained as part of a larger study carried out in six cities across the United States (see Kennedy, Caplan, & Piza, 2015).6 This presented a unique opportunity to investigate how place features influence crime across multiple environments at the same time in a consistent and standardized way. All data were sourced from the administrative records of po- lice departments, purchased from Infogroup7, or collected online from publicly available databases as shapefiles or XY coordinates.

The study settings included Chicago, Illinois, Kansas City, Mis- souri, and Newark, New Jersey. Chicago has the largest population with nearly 2.7 million people over a total land area of approxi- mately 227 square miles. Kansas City has the second largest pop- ulation, with about 460,000 people, and the largest land area of about 315 square miles. Finally, Newark has both the smallest population and land area with approximately 277,000 residents living within just 24 square miles. However, the population density in Newark (11,458 persons per square mile) is similar to Chicago (11,841 persons per square mile), where Kansas City is much less densely populated (1459 persons per square mile).

3.3. Outcome event

Outcome events were calendar year 2012 robbery incidents (see Table 1). Robberies are often a substantial source of fear among urban residents (Wright & Decker, 1997). Because over half involve aweapon, robberies have a high potential for serious harm (Federal Bureau of Investigation, 2012). The robbery rate in each study setting was three to six times higher than the national average (Federal Bureau of Investigation, 2012).

3.4. Potential risk factors

Various place features may aggravate or otherwise increase the risk for robbery. In total, 14 types of place features were selected in accordance with theory or existing empirical research to be tested for association with robbery in each study setting (see Table 1). Crime pattern theory describes how place features can act as crime generators by bringing together a large number of individuals for otherwise legitimate activity thereby increasing the potential for offenders to encounter ideal targets absent sufficient guardianship. Such features may include restaurants, retail shopping outlets, entertainment venues, commercial businesses, or public trans- portation stops (Brantingham & Brantingham, 1995). Conversely, place features may function as crime attractors, specifically draw- ing in individuals seeking to exploit well-known criminal oppor- tunities. Crime attractors may include bars, illicit markets, or insecure parking lots (Brantingham & Brantingham, 1995). Gener- ally speaking, classification of particular place features as crime generators or attractors is based on theoretical insights pertaining to the particular mechanisms that tie together those features and the specific crime problem at hand. Research in Chicago, forhttp://www.infogroup.com/

Table 1 Place feature counts and analysis parameters of risk terrain models for calendar year 2012 robbery incidents in Chicago, Kansas city, and Newark.

Variable Chicago Kansas City Newark

Outcome Event Robbery Incidents 13,480 1638 2001 Place Features Drug Arrests 3334 2521 4778 Parks 10,581 24,037 1350 Pawn Shops 68 14 36 Bars 1316 106 192 Foreclosures 15,305 311 779 Gas Stations 140 52 41 Grocery Stores 933 75 223 Health Centers & Gyms 176 34 11 Laundromats 173 28 37 Liquor Stores 926 258 87 Parking Stations 218 23 28 Schools 1021 171 124 Variety Stores 124 34 7 Bus Stops 10,711 3790 922

Model Parameters; Cell Size/Block Length (in feet): Chicago (213/426); Kansas City (231/462); Newark (226/452). Spatial Operationalization: Proximity and Density (Parks as Proximity; Drug Mar- kets as Density). Spatial Influence Extent/Analysis Increments: 3 Blocks/Half Blocks.

8 RTMDx can also create “protective” models, which search for negative spatial correlations between place features and crime. These features mitigate or otherwise decrease the risk of crime.

9 Arbitrary nodes were removed to ensure breaks in street segments only

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211206

example, examined the spatial dynamics of drug dealing and found that incidents of marijuana, heroin, crack, and cocaine dealing each exhibited unique set of place-based risk factors with variable spatial influence (Barnum et al., 2016).

Existing research provides guidance on specific place features that facilitate robbery. For example, an important factor in a rob- ber’s decision to offend is the presence or availability of cash or other goods that can be quickly converted into cash (St. Jean, 2007; Wright & Decker, 1997). Bernasco and Block (2011) examined numerous place features in Chicago that operate “small cash economies” where several individuals are likely to have cash on hand and found that several features, such as bars, liquor stores, grocery stores, gas stations, laundromats, liquor stores, pawnshops, and drug markets, among others, were related to robbery. Wright and Decker (1997) and St. Jean (2007), based on interviews with actual robbers, explain how these types of place features may attract robbers for additional reasons. For example, bars or liquor stores contain intoxicated patrons who are particularly vulnerable because they are not in a position to offer resistance (see also Groff, 2014; Roncek& Bell, 1981; Roncek&Maier,1991). Pawnshops allow robbers to easily and quickly convert stolen goods into cash. Moreover, they could provide robbers with easy access to weapons to commit their offense. Drug markets may facilitate robbery because drug buyers and sellers, owing to their extralegal activities, are not likely to report their victimization.

Other important elements that inform the decision calculus of robbers relate to the accessibility and familiarity of places. In other words, robbers prefer places that are easy to get to and offer quick escape once they have committed their robbery (St. Jean, 2007; Wright & Decker, 1997). This is likely to include places with fea- tures such as grocery stores, gas stations, laundromats, or other types of retail and commercial stores (Bernasco & Block, 2011; Smith, Frazee, & Davison, 2000). These types of businesses are likely to be located in popular areas that are frequented by a large number of people. They are convenient to travel to, owing to their connectedness to the rest of the city to enhance people traffic to increase business. They are likely to be located near parking loca- tions or public transportation stops, other features that have been associated with crime, and specifically robbery (Bernasco & Block,

2011; Groff & Lockwood, 2014; Hart & Miethe, 2014; Smith et al., 2000). These features also reduce the risk for detection because they have an ongoing rhythm of activity in which to easily blend in (St. Jean, 2007; Wright & Decker, 1997).

Other features may create suitable conditions for robbery because they lack sufficient guardianship and thereby enhance the probability that a robber will successfully offend and remain anonymous. For example, Groff and McCord (2012) found that parks increased crimes in the surrounding areas, particularly when they had characteristics that reduced guardianship. Because they are unattended, foreclosures may also make robbery more likely. Spelman (1993) found that crime was higher near abandoned properties; a number of additional studies have documented a link specifically between foreclosures and violent crime (Baumer, Wolff, & Arnio, 2012; Bess, 2008; Ellen, Lacoe, & Sharygin, 2013). Several studies have found schools to be associated with higher levels of violent crime (Bernasco & Block, 2009; Roncek & Faggiani, 1985; Roncek & Lobosco, 1983). Bernasco and Block (2009) suggest that schools provide good targets, but schools themselves may have high surveillance. However, this might not be the case in the sur- rounding areas, or after hours, where individuals may congregate unsupervised. Finally, health centers and gyms have received little attention in prior research, but may provide ideal targets for rob- bers. For example, these features may be attractive to robbers because they are often open around the clock and unstaffed (and consequently unguarded). People visit the gyms late at night to avoid crowds, often carrying only their wallet or a small electronic device such as an iPhone or iPod that can be easily grabbed as someone is coming or going to the gym and then converted into cash (e.g., see Caplan & Kennedy, 2016, p. 99). Also, gyms are often located in highly trafficked areas owing to the need to be easily accessible for people on the way to and fromwork, school, or other daily activities.

3.5. Model parameters and operationalization

The Risk Terrain Modeling Diagnostics Utility (RTMDx; Caplan& Kennedy, 2013) was used to create RTMs for calendar year 2012 robbery incidents and examine risky place features and their spatial influence in Chicago, Newark, and Kansas City. Besides the study setting boundaries, all other parameters were standardized across models to generate results that reflected variation in each jurisdiction’s environmental backcloth rather than variations in model parameters or methods of testing.

First, “aggravating” models were specified, which identify pos- itive spatial associations between potential risk factors and the outcome event, to determine risky place features8. Next, “cell sizes” and “block lengths,” which served as the units of analysis for each model, were specified as one-half the average block length and the average block length in each study setting, respectively (Caplan, Kennedy, & Piza, 2013). This is consistent with research suggest- ing the importance of examining the dynamics of crime at the micro unit of analysis (Weisburd, Bernasco,& Bruinsma, 2008). Cell sizes and block lengths were determined via street centerline files9

(see Table 1). The next set of parameters included “spatial operationalization,”

“maximum spatial influence,” and “analysis increments” for each place feature tested. Spatial operationalization describes the

occurred at true intersections.

Table 2 Risk terrain model results for calendar year 2012 robbery incidents in Chicago, Kansas city, and Newark.

Risk factor Chicago Kansas City Newark

O/SI RRV O/SI RRV O/SI RRV

Foreclosures P/852 4.51 P/1386 1.68 P/1356 9.61 Gas Stations P/213 4.60 D/462 2.11 P/226 2.65 Grocery Stores P/1065 1.57 P/1386 1.73 D/1356 1.47 Health Centers & Gyms e e e e e e Laundromats P/213 2.27 e e P/226 2.89 Parking Stations P/213 1.96 e e P/904 1.53 Variety Stores P/1278 1.25 D/1386 1.64 e e Bus Stops D/426 2.55 D/1155 5.38 P/226 3.68 Bars P/213 1.83 e e P/678 1.46 Drug Markets D/1065 2.36 D/231 8.69 D/226 2.39 Schools P/1278 1.39 e e P/1356 1.57 Parks e e P/1386 1.57 e e Liquor Stores P/213 2.97 P/1386 2.19 P/1356 1.50 Pawn Shops P/1278 1.29 e e e e

Note: O: Operationalization (P¼Proximity or D ¼ Density); SI: Spatial Influence (in feet) RRV: Relative Risk Value.

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211 207

particular influence of each feature; that is, crime risk can be higher at places as a function of proximity to features or at places with a dense concentration or clustering of features. Theoretically, either, both, or neither operationalization of features could be true, given the particular environment (for a detailed discussion of oper- ationalizing spatial influence, see Caplan, 2011). To determine when features generate themost risk, the spatial operationalization of most features was tested as both proximity and density10. There were two exceptions. First, parks were represented as polygon shapefiles, but RTMDx only accepts point features. Therefore, park polygons were converted to a representative grid of points and tested as proximity only. Second, drug arrest incidents11 represent a “fleeting” phenomenon that may reflect a standing quality of the environment when concentrated at places (Caplan et al., 2013) and tested as density only.

Finally, research suggests that the spatial influence of place features is geographically limited and decays with distance (e.g., Groff & Lockwood, 2014). Therefore, the spatial influence of each feature was tested to a maximum extent of three blocks at half- block analysis increments. In sum, the spatial influence of each place feature was tested at one-half block; one block; one-and-one- half a block; two blocks; two-and-one-half blocks, and three blocks, as a function of proximity and density, respectively.

3.6. Statistical analysis

The statistics of RTMDx are only briefly described here (for a detailed discussion, see Heffner, 2013, p. 35). RTMDx begins by building an elastic net penalized regression model assuming a Poisson distribution of events. Cross-validation is used to reduce the initial set of variables.12RTMDx continues by building additional models, one assuming a Poisson distribution and the other a negative binomial distribution. The Bayesian Information Criteria (BIC) is measured for a null model and then again as each new variable is added. RTMDx continues in this iterative fashion until the addition or removal of a new variable does not allow the model to surpass the BIC score of the previous candidate model. RTMDx finishes by choosing the model with the lowest BIC score between the two distributions. Coefficients for the remaining variables (i.e., risk factors) are rescaled between theminimum andmaximum risk values to produce relative risk values (RRV),13 weighting each factor relative to one another. Relative risk scores (RRS) are computed for each cell within the study setting.

4. Results

Table 2 presents the risk factors for robbery in each study setting. For each risk factor, the models identified the optimal spatial operationalization, the maximum spatial influence, and a RRV. Of the 14 types of place features tested,12 were risk factors for robbery in Chicago; 8 in Kansas City; and 10 in Newark. RRVs ranged from 1.25 to 4.60 in Chicago, from 1.57 to 8.69 in Kansas City, and from 1.46 to 9.61 in Newark. Gas stations were the riskiest

10 RTMDx measures proximity using Euclidean (i.e., straight-line) distance. Kernel density estimation (KDE) is used to measure density. 11 Following previous studies, drug arrest incidents are utilized to represent drug markets (e.g., see Eck, 1995; Rengert, Ratcliffe, & Chakravorty, 2005; Weisburd & Green, 1995). 12 Spatial autocorrelation can be problematic to the extent that when present, significance values may be affected and Type I errors are more likely. However, RTMDx employ cross-validation, which deemphasizes significance testing for var- iable selection (see Heffner, 2013: 38). 13 Relative risk values are calculated by exponentiating risk factor coefficients provided by the risk terrain model.

factor in Chicago. Places located within 213 feet (i.e., one-half a block) of a gas stationwere at 4.60more risky for robbery compared to places absent the spatial influence of any risk factors. In Kansas City, the riskiest factor was drug markets and places where drug arrest incidents clustered within a 231-foot (i.e., one-half block) bandwidth were 8.69 times more risky for robbery compared to places absent any risk factors’ influence. Foreclosures were the riskiest feature in Newark and places within 1356 feet (i.e., three blocks) of them were 9.61 as risky for robbery compared to places where no risk factors’ influence was present.

Six features, including foreclosures, gas stations, grocery stores, bus stops, drug markets, and liquor stores were risk factors for robbery in all three study settings. This is largely consistent with findings reported by other studies (e.g., Bernasco & Block, 2011; Hart & Miethe, 2014; Smith et al., 2000; Spelman, 1993; St. Jean, 2007; Wright & Decker, 1997). Some commonly identified corre- lates of robbery (as reported in aforementioned studies), however, were not risk factors in one or more study settings despite theo- retical expectations or extant research suggesting they would have been. For example, parks were not risk factors for robbery in Chi- cago; laundromats, parking stations, bars, schools, and pawn shops were not risk factors in Kansas City; and variety stores, parks, and pawn shops were not risk factors in Newark. Health centers and gyms were not risk factors for robbery in any study setting.

Variations were observed in risk factors’ spatial influences, specifically, their operationalizations, extents of influence, and weights of influence. With regard to operationalization, risk could be higher for robbery near features or at places where features clustered. Although foreclosures, gas stations, grocery stores, bus stops, and liquor stores were risk factors for robbery in all three study settings, only foreclosures and liquor stores shared the same operationalization. In all three study settings, risk was higher near foreclosures and liquor stores. Operationalizations varied for gas stations, grocery stores, and bus stops. For example, risk was also higher near gas stations in Chicago and Newark, and near grocery stores in Chicago and Kansas City. On the hand, risk was higher at places where gas stations clustered in Kansas City and at places where grocery stores clustered in Newark. Similarly, risk was higher at places where bus stops clustered in Chicago and Kansas City, but near bus stops in Newark. Drug markets was tested only as a function of density and was therefore operationalized as such in all three study settings.

The spatial influence of each feature was tested to a maximum extent of 3 blocks in half-block increments and the results show

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211208

that the risk factors for robbery affected a different range of space across the three study settings. For example, although the spatial influence of liquor stores extended to just half a block in Chicago, it reached a full three blocks in Kansas City and Newark. Likewise, the spatial influence of drug markets in Chicago extended 2.5 blocks, but was limited to just half a block in Kansas City and Newark. Such variations could be observed for other risk factors between study settings, such as parking stations, bus stops, and bars. However, other risk factors, such as foreclosures, gas stations, grocery stores, Laundromats, variety stores, and schools, had similar extents of spatial influence.

Finally, some risk factors aggravated robbery more in certain study settings compared to others. This was measured as a function of each risk factors’ RRV, essentially weights of influence. This was most salient for foreclosures, which were much riskier for robbery in Newark (RRV ¼ 9.61) than Chicago (RRV ¼ 4.51) or Kansas City (RRV ¼ 1.68). Moreover, gas stations generated over twice as much risk for robbery in Kansas City (RRV ¼ 2.11) and Newark (RRV ¼ 2.65), but nearly five times as much risk in Chicago (RRV ¼ 4.60). Drug markets were nearly nine times as risky for robbery in Kansas City (RRV ¼ 8.69), but just over twice as risky in Chicago (RRV ¼ 2.36) and Newark (RRV ¼ 2.39). Finally, bus stops were over twice as risky for robbery in Chicago (RRV ¼ 2.55), over three times as risky in Newark (RRV ¼ 3.68), and over five times as risky in Kansas City (RRV ¼ 5.38). Although the same risk factors were oftentimes more problematic in one setting compared to another, some consistencies emerged. For example, though signif- icant aggravators for robbery, grocery stores, parking stations, bars, and schools were generally less problematic than other place features.

5. Discussion

RTM, a geospatial crime forecasting and diagnostic method, was utilized to identify risk factors for robbery and their spatial influ- ence for comparison in Chicago, Newark, and Kansas City. The goal was to examine the physical landscape of different urban envi- ronments and their relative influence on crime. As environmental theories of crime would suggest, it was assumed that certain place features would increase the risk for robbery. However, it was hy- pothesized that these features and their particular influence on robbery would vary from one environment to the next given their unique patterning throughout each jurisdiction’s environmental backcloth. Upon peering through each environment’s crime risk kaleidoscope, a number of interesting findings emerged. First, each study setting had a unique set of risk factors for robbery. A second important finding was that the risk factors were oftentimes more problematic in some study settings as compared to others. Third, the spatial influence of risk factors oftentimes varied across study settings. In other words, risk was higher near features in some settings and at places where features clustered in other settings. Moreover, while the spatial influence of certain features extended several blocks in some settings it was relatively limited in others. Collectively, these findings suggest a unique set of underlying spatial dynamics that influence the emergence and persistence of robbery at places within each jurisdiction. Crime pattern theory (Brantingham & Brantingham, 2008) suggests that these divergent spatial dynamics are the product of nuances in the physical aspects of each jurisdiction’s environmental backcloth. As human creations and artifacts of more localized social, cultural, legal and economic processes, built environments are unique in their forms and func- tions. The ways in which physical landscapes shape opportunities for crime is not necessarily the same from one environment to the next.

This is reflected by differences in risk factors for robbery and

their spatial influence across jurisdictions. Although it is not possible to speculate about all the variations observed in the findings, more broad theoretical considerations are warranted. For example, risk was higher near features in some jurisdictions but higher at places where the same types of features clustered in other jurisdictions. One possible explanation is that risk as a result of proximity to features may be reflective of the feature’s crime attracting properties (Brantingham & Brantingham, 1995). In other words, proximity to gas stations in Chicago and Newark could suggest that in these jurisdictions, gas stations have specific qual- ities that make robbery more likely nearby, such as being open late, providing a place for people to hang out (if there is a convenience store that sells cheap food and drink), and drawing a regular supply of customers who have cash on hand that make good targets. Conversely, risk at places due to the clustering of features may be more reflective of the feature’s crime generating properties (Brantingham & Brantingham, 1995). Perhaps in a large, relatively dispersed jurisdiction such as Kansas City, many gas stations are otherwise isolated in low traffic areas so proximity to just any gas station is not very risky. However, places where gas stations are clustered may be high activity nodes whereby both offenders and targets are concentrated and interact more frequently to increase the chances that good opportunities for crime will arise.

The distribution of place features along certain paths and rela- tive to various types of edges may help to explain differences in extents of spatial influence too (Brantingham & Brantingham, 1993). For example, paths consist of the underlying network of roads, sidewalks, or other routes of travel. Place features commonly located along more complex path networks may allow for a greater extent of influence as these networks have a greater capacity to support the everyday travel of offenders and victims in nearby areas. Conversely, robbery is more likely to happen at or very close to features that are located on a single, isolated road (e.g., see Beavon, Brantingham, & Brantingham, 1994; Davies & Johnson, 2015). Edges, or physical and perceptual barriers in the environ- ment, could also play an important role in shaping the extent of influence of risk factors. For example, features near strong and clear edges are likely to host crimes within a very small area compared to features near “fuzzy” edges, which disperse crime over a large area (Brantingham, Brantingham, Vajihollahi, & Wuschke, 2009). Addi- tional research is required to directly incorporate path networks and the various edges throughout the environment backcloth to better understand how these elements work in conjunction to shape the influence of the physical landscape on crime.

Another important consideration is that the spatial arrangement of place features relative to one another throughout the environ- ment may affect how they collocate and interact or work together to generate robbery risk. When certain features converge at places it creates a dynamic that is conducive to illegal behavior. Certain features, such as drug markets, may increase the risk for robbery because they operate primarily in cash and lack a formal third party to resolve disputes among competing dealers and their customers. Thus, robbery is primarily the result of participants in the drug market targeting one another. However, when collocated with or nearby other features, such as gas stations or public transportation stops, the criminogenic mechanisms may be enhanced because the latter features draw a larger number of people into the area, enabling a new set of suitable victims. For example, drug users may begin to target people who do not participate in the drug markets for their cash or valuables to purchase drugs. Thus, place features may increase robbery risk in their own right, but may become exponentially riskier when located near other features with crim- inogenic qualities. Indeed, features themselves may only become risky as a result of such co-effects. This study did not explicitly test the interaction effects between the risk factors to examine how the

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211 209

influence of certain features shapes that of other nearby features, but this is an important avenue for future inquiry.

At the same time, the distribution of place features throughout an environment may explain why certain features are not risky at all. Some place features typically associated with robbery may be located near “protective” features that undermine crime potential through enhanced guardianship (Clarke & Eck, 2005). For example, studies have found that vacant lot cleaning and greening reduces crime and fear of crime (Branas, Cheney, MacDonald, Tam, Jackson, & Ten Have, 2011; Garvin, Cannuscio, & Branas, 2012). As a result, clean and green vacant lots might function as a crime mitigator by increasing the levels of residents’ activities in the community. Another study found that levels of violence were lower in areas with more churches per capita (Lee, 2006), suggesting that churches may be another place feature with high levels of guard- ianship that may negate the aggravating mechanisms of nearby features (see also Eck & Weisburd, 1995). Additional research is necessary to examine whether certain features do indeed provide a protective effect against crime, and moreover, how protective fea- tures interact with nearby risk factors.

These findings should be interpreted cautiously in light of a number of important limitations. First, and as with any study, these findings are directly tied to the quality of the data utilized in the analyses. With regard to the outcome event, this study utilized police recorded incidents of robbery. Although prior studies have utilized robbery incident data (e.g., Braga et al., 2011), they could be limited due to individuals’ failing to report crimes or police de- cisions not to record crimes. Another issue was that it was not possible to make potentially relevant distinctions in the outcome event. For example, it is possible that the different findings would have emerged by distinguishing street and commercial robbery. Data quality is an equally important issue for the independent variables, or in this case potential risk factors for robbery, because improper classification or missing data can lead to model mis- specification. Most place feature data utilized in this study were purchased from Infogroup. However, some place feature data were obtained from police departments or from open data portals. This could have introduced inconsistencies across jurisdictions. Although the place feature data utilized in the studywere inspected for accuracy and completeness, it was not possible to visit the location of every place feature in each jurisdiction for true verification.

A second limitation was that the analyses here were exclusively concerned with the effects of place features on crime. Although place features do influence crime, it is likely that other variables are relevant, such as the broader community structural forces of poverty, population heterogeneity, neighborhood instability, and collective efficacy (Sampson& Groves, 1989; Sampson, Morenoff, & Gannon-Rowley, 2002). This is an important limitation, but it is worthwhile to note that a number of studies have examined the effects of place features on crime while controlling for community structure. For example, Groff and Lockwood (2014) controlled for population size and heterogeneity, disadvantage, and neighbor- hood stability and found that certain features, such as bars and public transportation stops, remain important predictors of violent crime. In another study Drawve, Thomas, and Walker (2016) uti- lized RTM to identify risk factors for violent crime and then created an aggregate neighborhood risk of crime measure (ANROC). They found that the ANROC measure was a significant and important predictor of violent crime, even when controlling for concentrated disadvantage and neighborhood stability. These studies demon- strate that place features can play an important role in crime, above and beyond community structure. However, it is likely that both environmental and social variables work together in complex ways to produce illegal behavior. For example, a recent study by Piza,

Feng, Kennedy, and Caplan (2016) found that the spatial influence of risk factors for motor vehicle theft and motor vehicle recovery varied across different neighborhood contexts. More specifically, they determined that certain place features increased the risk of motor vehicle theft and recovery, but that the criminogenic influ- ence of these factors was either heightened or mitigated by certain neighborhood dynamics. They concluded that theories of envi- ronmental criminology and social disorganization are comple- mentary perspectives. Although these studies add to the validity of the current study, future research that examines differences in risk factors for crime and their spatial influence should seek to incor- porate community structure, as well as other possible sources of variation (e.g., policing intensity).

That this study did not include a temporal component was a third limitation because research has shown that risk factors’ spatial influences can vary greatly across time periods (Irvin- Erickson, 2014). Finally, risk factors for robbery were compared across urban environments. It is less clear if similar findings would hold in smaller, more homogenous, suburban and rural settings. Future work that overcomes these limitations would greatly enhance the validity of the current findings and provide a better understanding of the kaleidoscope of risky places for crime.

Keeping in mind these limitations, this study has important policy implications. For example high crime places, also referred to as “hot spots” given their relatively high density of crime relative to other places throughout the broader jurisdiction (Sherman, 1995), provide an important avenue for crime prevention. It is well- documented in the criminal justice literature that police operate more efficiently and effectively when focusing their resources and efforts on high crime places (Braga, Papachristos, & Hureau, 2012). However, whereas hot spots inform police about places that are exposed to crime, they do less to describe the particular qualities of the places that make them vulnerable to crime (Kennedy et al., 2015). The methods utilized in this study can be incorporated into regular police practice to provide guidance about what exactly police should focus on at high crime places (see also Kennedy et al., 2011). This is particularly important given the current finding that the characteristics of high crime places depend upon the jurisdic- tion under consideration.

Police agencies can view specific crime problems through their own crime risk kaleidoscope to better understand and diagnose local crime vulnerabilities with appropriately customized in- terventions. This may include deploying commonpolice tactics that have been found to be effective in addressing high crime places, such as directed patrol (Koper, Taylor, & Woods, 2013; Rosenfeld, Deckard, & Blackburn, 2014; Sherman & Weisburd, 1995) or foot patrol (Novak, Fox, Carr, & Spade, 2016; Piza & O’Hara, 2014; Ratcliffe, Taniguchi, & Wood, 2011), but with an even more refined focus on the risk factors that are present. For example, while on patrol, officers may direct their attention towards specific risk factors, whether that includes reporting unsecured vacant prop- erties to other municipal agencies, approaching people at public transportation stops to educate them of recent crime problems and good ways to stay safe, or increasing their visibility and presence around bars. What is important is police can identify and prioritize actual crime vulnerabilities.

Tactics like directed patrol and foot patrol are important crime prevention measures, but more holistic long-term strategies, such as problem-oriented policing (Goldstein, 1990) or situational crime prevention (Clarke, 1980), are likely to be the most effective to this end (Skogan & Frydl, 2004). The ultimate goal of any place-based police strategy is to fundamentally change the characters of pla- ces that are problematic (Braga& Clarke, 2014). RTM can provide an empirical basis to problem assessments and inform the develop- ment of tailored interventions to address the localized conditions

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211210

that affect crime (see Caplan & Kennedy, 2016). For example, Kennedy, Caplan, and Piza (2015) examined the crime prevention value of various police activities that were specifically designed to mitigate or eliminate the criminogenic spatial influence of crime risk factors in multiple jurisdictions across the United States. They found that by employing risk reduction initiatives, measurable re- ductions in crime could be achieved. In one study setting, robbery was reduced by 42% in target areas relative to control areas. By addressing the environments that foster opportunities for crime, it is more likely that any crime prevention gains can be sustained over time.

Funding

This research was supported in part by a grant from the National Institute of Justice (2012-IJ-CX-0038). The views presented are those of the authors and do not necessarily represent the position of the National Institute of Justice.

References

Andresen, M. A. (2014). Environmental criminology: Evolution, theory, and practice. New York, NY: Routledge.

Baumer, E. P., Wolff, K. T., & Arnio, A. N. (2012). A multicity neighborhood analysis of foreclosure and crime. Social Science Quarterly, 93, 577e601.

Barnum, Jeremy D., Campbell, Walter L., Trocchio, Sarah, Caplan, Joel M., & Kennedy, Leslie W. (May 23 2016). Examining the Environmental Characteristics of Drug Dealing Locations. Crime & Delinquency, Sage Journals.

Beavon, D. J. K., Brantingham, P. L., & Brantingham, P. J. (1994). The influence of street networks on the patterning of property offenses. Crime Prevention Studies, 2, 115e148.

Bernasco, W., & Block, R. (2009). Where offenders choose to attack: A discrete choice model of robberies in Chicago. Criminology, 47, 93e130.

Bernasco, W., & Block, R. (2011). Robberies in Chicago: A block-level analysis of the influence of crime generators, crime attractors, and offender anchor points. Journal of Research in Crime and Delinquency, 48, 33e57.

Bess, M. (2008). Assessing the impact of home foreclosures in Charlotte neighbor- hoods. Geography and Public Safety. A Quarterly Bulletin of Applied Geography for the Study of Crime and Public Safety, 1(3), 2e4.

Braga, A. A., & Clarke, R. V. (2014). Explaining high-risk concentrations of crime in the city: Social disorganization, crime opportunities, and important next steps. Journal of Research in Crime and Delinquency, 51(4), 480e498.

Braga, A. A., Hureau, D. M., & Papachristos, A. V. (2011). The relevance of micro places to citywide robbery trends: A longitudinal analysis of robbery incidents at street corners and block faces in Boston. Journal of Research in Crime and Delinquency, 48(1), 7e32.

Braga, A. A., Papachristos, A. V., & Hureau, D. M. (2010). The concentration and stability of gun violence at micro places in Boston, 1980e2008. Journal of Quantitative Criminology, 26(1), 33e53.

Braga, A., Papachristos, A., & Hureau, D. (2012). Hot spots policing effects on crime. Campbell Systematic Reviews, 8, 1e96.

Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Have, T., et al. (2011). A difference-in-differences analysis of health, safety, and greening vacant urban space. American Journal of Epidemiology, 174(11), 1296e1306.

Brantingham, P., & Brantingham, P. (1984). Patterns in crime. New York, NY: Mac- Millin Publishing Company.

Brantingham, P. L., & Brantingham, P. J. (1993). Nodes, paths and edges: Consider- ations on the complexity of crime and the physical environment. Journal of Environmental Psychology, 13(1), 3e28.

Brantingham, P., & Brantingham, P. (1995). Criminality of place. European Journal on Criminal Policy and Research, 3, 5e26.

Brantingham, P., & Brantingham, P. (2008). Crime pattern theory. In R. Wortley, & L. Mazerolle (Eds.), Environmental criminology and crime analysis. New York, NY: Routledge.

Brantingham, P. L., Brantingham, P. J., Vajihollahi, M., & Wuschke, K. (2009). Crime analysis at multiple scales of aggregation: A topological approach. In D. Weisburd, W. Bernasco, & G. J. N. Bruinsma (Eds.), Putting crime in its Place: Units of analysis in geographic criminology. New York: NY: Springer.

Burgess, E. W. 1984(1925, 1967). The growth of the city: An introduction to a research project. In R. Park, & E. Burgess (Eds.), The city: Suggestions for inves- tigation of human behavior in the urban environment, 1984(1925, 1967). Chicago, IL: University of Chicago Press.

Bursik, R. J. (1988). Social disorganization and theories of crime and delinquency: Problems and prospects. Criminology, 26(4), 519e552.

Caplan, J. M. (2011). Mapping the spatial influence of crime correlates: A compar- ison of operationalization schemes and implications for crime analysis and criminal justice practice. Cityscape, 13, 57e83.

Caplan, J. M., & Kennedy, L. W. (2013). Risk terrain modeling diagnostics utility

(Version 1.0). Newark, NJ: Rutgers Center on Public Security. Caplan, J. M., & Kennedy, L. W. (2016). Risk terrain modeling: Crime prediction and risk

reduction. Oakland, CA: University of California Press. Caplan, J. M., Kennedy, L. W., & Miller, J. (2011). Risk terrain modeling: Brokering

criminological theory and GIS methods for crime forecasting. Justice Quarterly, 28, 360e381.

Caplan, J. M., Kennedy, L. W., & Piza, E. L. (2013). Risk terrain modeling diagnostics utility user manual (Version 1.0). Newark, NJ: Rutgers Center on Public Security.

Clarke, R. V. (1980). Situational crime prevention: Theory and practice. British Journal of Criminology, 20, 136e147.

Clarke, R. V., & Cornish, D. B. (1985). Modeling offenders’ decisions: A framework for research and policy. Crime and Justice, 6, 147e185.

Clarke, R. V., & Eck, J. E. (2005). Crime analysis for problem solvers in 60 small steps. Washington, D.C.: U.S. Department of Justice, Office of Community Oriented Policing Services.

Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American sociological review, 44, 588e608.

Davies, T., & Johnson, S. D. (2015). Examining the relationship between road structure and burglary risk via quantitative network analysis. Journal of Quan- titative Criminology, 31(3), 481e507.

Drawve, G., Thomas, S. A., & Walker, J. T. (2016). Bringing the physical environment back into neighborhood research: The utility of RTM for developing an aggre- gate neighborhood risk of crime measure. Journal of Criminal Justice, 44, 21e29.

Eck, J. E. (1995). A general model of the geography of illicit retail marketplaces. In J. E. Eck, & D. Weisburd (Eds.), Crime and place: Crime prevention studies (Vol. 4, pp. 67e95). Monsey, NY: Criminal Justice Press.

Eck, J. E., & Weisburd, D. (1995). Crime places in crime theory. Crime and Place, Crime Prevention Studies, 4(1), 1e33.

Ellen, I. G., Lacoe, J., & Sharygin, C. A. (2013). Do foreclosures cause crime? Journal of Urban Economics, 74, 59e70.

Federal Bureau of Investigation. (2012). Crime in the United States. Washington, DC: Department of Justice.

Garvin, E., Cannuscio, C., & Branas, C. (2012). Greening vacant lots to reduce violent crime: A randomised controlled trial. Injury Prevention, 19(3), 198e203.

Goldstein, H. (1990). Problem-oriented policing. Philadelphia, PA: Temple University Press.

Groff, E. R. (2014). Quantifying the exposure of street segments to drinking places nearby. Journal of Quantitative Criminology, 30, 527e548.

Groff, E. R., & Lockwood, B. (2014). Criminogenic facilities and crime across street segments in Philadelphia: Uncovering evidence about the spatial extent of fa- cility influence. Journal of Research in Crime and Delinquency, 51, 277e314.

Groff, E. R., & McCord, E. S. (2012). The role of neighborhood parks as crime gen- erators. Security Journal, 25(1), 1e24.

Harris, C. D., & Ullman, E. L. (1945). The nature of cities. The Annals of the American Academy of Political and Social Science, 242, 7e17.

Hart, T. C., & Miethe, T. D. (2014). Street robbery and public bus stops: A case study of activity nodes and situational risk. Security Journal, 27, 180e193.

Heffner, J. (2013). Statistics of the RTMDx utility. In J. M. Caplan, L. W. Kennedy, & E. L. Piza (Eds.), Risk terrain modeling diagnostics utility user manual, Version 1.0. Newark, NJ: Rutgers Center on Public Security.

Hoyt, H. (1939). The structure and growth of residential neighborhoods in American cities. Washington, DC: Federal Housing Administration.

Irvin-Erickson, Y. (2014). Identifying risky places for crime: An analysis of the crimi- nogenic spatiotemporal influences of landscape features on street robberies (Doctoral Dissertation). Newark, NJ: Rutgers University-Graduate School.

Kennedy, L. W. (1983). The urban kaleidoscope: Canadian perspectives. New York, NY: McGraw-Hill Ryerson Limited.

Kennedy, L. W., Caplan, J. M., & Piza, E. L. (2011). Risk clusters, hotspots, and spatial intelligence: Risk terrain modeling as an algorithm for police resource alloca- tion strategies. Journal of Quantitative Criminology, 27, 339e362.

Kennedy, L. W., Caplan, J. M., & Piza, E. L. (2015). A multi-jurisdictional test of risk terrain modeling and place-based evaluation of environmental risk-based patrol deployment strategies. Newark, NJ: Rutgers Center on Public Security.

Kennedy, L. W., Caplan, J. M., Piza, E. L., & Buccine-Schraeder. (2015). Vulnerability and exposure to crime: Applying risk terrain modeling to the study of assault in Chicago. Applied Spatial Analysis. http://dx.doi.org/10.1007/s12061-015-9165-z.

Koper, C. S., Taylor, B. G., & Woods, D. J. (2013). A randomized test of initial and residual deterrence from directed patrols and use of license plate readers at crime hot spots. Journal of Experimental Criminology, 9(2), 213e244.

Lee, M. R. (2006). The religious institutional base and violent crime in rural areas. Journal for the Scientific Study of Religion, 45(3), 309e324.

Lynch, K. (1960). The image of the city. Cambridge, MA: The MIT Press. Novak, K. J., Fox, A. M., Carr, C. M., & Spade, D. A. (2016). The efficacy of foot patrol in

violent places. Journal of Experimental Criminology, 12(3), 465e475. Piza, E., Feng, S., Kennedy, L., & Caplan, J. (2016). Place-based correlates of motor

vehicle theft and recovery: Measuring spatial influence across neighborhood context. Urban Studies. http://dx.doi.org/10.1177/0042098016664299.

Piza, E. L., & O’Hara, B. A. (2014). Saturation foot-patrol in a high-violence area: A quasi- experimental evaluation. Justice Quarterly, 31(4), 693e718.

Poon, L. (2015, September 11). Mapping the ‘urban fingerprints’ of cities. CityLab. Retrieved from: http://www.citylab.com/housing/2015/09/mapping-the-urban- fingerprints-of-cities/404923/.

Ratcliffe, J. H., Taniguchi, T., Groff, E. R., & Wood, J. D. (2011). The philadelphia foot patrol experiment: A randomized controlled trial of police patrol effectiveness in violent crime hotspots. Criminology, 49(3), 795e831.http://refhub.elsevier.com/S0143-6228(16)30828-1/sref1http://refhub.elsevier.com/S0143-6228(16)30828-1/sref1http://refhub.elsevier.com/S0143-6228(16)30828-1/sref2http://refhub.elsevier.com/S0143-6228(16)30828-1/sref2http://refhub.elsevier.com/S0143-6228(16)30828-1/sref2http://refhub.elsevier.com/S0143-6228(16)30828-1/sref76http://refhub.elsevier.com/S0143-6228(16)30828-1/sref76http://refhub.elsevier.com/S0143-6228(16)30828-1/sref76http://refhub.elsevier.com/S0143-6228(16)30828-1/sref76http://refhub.elsevier.com/S0143-6228(16)30828-1/sref3http://refhub.elsevier.com/S0143-6228(16)30828-1/sref3http://refhub.elsevier.com/S0143-6228(16)30828-1/sref3http://refhub.elsevier.com/S0143-6228(16)30828-1/sref3http://refhub.elsevier.com/S0143-6228(16)30828-1/sref4http://refhub.elsevier.com/S0143-6228(16)30828-1/sref4http://refhub.elsevier.com/S0143-6228(16)30828-1/sref4http://refhub.elsevier.com/S0143-6228(16)30828-1/sref5http://refhub.elsevier.com/S0143-6228(16)30828-1/sref5http://refhub.elsevier.com/S0143-6228(16)30828-1/sref5http://refhub.elsevier.com/S0143-6228(16)30828-1/sref5http://refhub.elsevier.com/S0143-6228(16)30828-1/sref6http://refhub.elsevier.com/S0143-6228(16)30828-1/sref6http://refhub.elsevier.com/S0143-6228(16)30828-1/sref6http://refhub.elsevier.com/S0143-6228(16)30828-1/sref6http://refhub.elsevier.com/S0143-6228(16)30828-1/sref7http://refhub.elsevier.com/S0143-6228(16)30828-1/sref7http://refhub.elsevier.com/S0143-6228(16)30828-1/sref7http://refhub.elsevier.com/S0143-6228(16)30828-1/sref7http://refhub.elsevier.com/S0143-6228(16)30828-1/sref8http://refhub.elsevier.com/S0143-6228(16)30828-1/sref8http://refhub.elsevier.com/S0143-6228(16)30828-1/sref8http://refhub.elsevier.com/S0143-6228(16)30828-1/sref8http://refhub.elsevier.com/S0143-6228(16)30828-1/sref8http://refhub.elsevier.com/S0143-6228(16)30828-1/sref9http://refhub.elsevier.com/S0143-6228(16)30828-1/sref9http://refhub.elsevier.com/S0143-6228(16)30828-1/sref9http://refhub.elsevier.com/S0143-6228(16)30828-1/sref9http://refhub.elsevier.com/S0143-6228(16)30828-1/sref9http://refhub.elsevier.com/S0143-6228(16)30828-1/sref10http://refhub.elsevier.com/S0143-6228(16)30828-1/sref10http://refhub.elsevier.com/S0143-6228(16)30828-1/sref10http://refhub.elsevier.com/S0143-6228(16)30828-1/sref12http://refhub.elsevier.com/S0143-6228(16)30828-1/sref12http://refhub.elsevier.com/S0143-6228(16)30828-1/sref12http://refhub.elsevier.com/S0143-6228(16)30828-1/sref12http://refhub.elsevier.com/S0143-6228(16)30828-1/sref13http://refhub.elsevier.com/S0143-6228(16)30828-1/sref13http://refhub.elsevier.com/S0143-6228(16)30828-1/sref14http://refhub.elsevier.com/S0143-6228(16)30828-1/sref14http://refhub.elsevier.com/S0143-6228(16)30828-1/sref14http://refhub.elsevier.com/S0143-6228(16)30828-1/sref14http://refhub.elsevier.com/S0143-6228(16)30828-1/sref15http://refhub.elsevier.com/S0143-6228(16)30828-1/sref15http://refhub.elsevier.com/S0143-6228(16)30828-1/sref15http://refhub.elsevier.com/S0143-6228(16)30828-1/sref16http://refhub.elsevier.com/S0143-6228(16)30828-1/sref16http://refhub.elsevier.com/S0143-6228(16)30828-1/sref16http://refhub.elsevier.com/S0143-6228(16)30828-1/sref17http://refhub.elsevier.com/S0143-6228(16)30828-1/sref17http://refhub.elsevier.com/S0143-6228(16)30828-1/sref17http://refhub.elsevier.com/S0143-6228(16)30828-1/sref17http://refhub.elsevier.com/S0143-6228(16)30828-1/sref19http://refhub.elsevier.com/S0143-6228(16)30828-1/sref19http://refhub.elsevier.com/S0143-6228(16)30828-1/sref19http://refhub.elsevier.com/S0143-6228(16)30828-1/sref20http://refhub.elsevier.com/S0143-6228(16)30828-1/sref20http://refhub.elsevier.com/S0143-6228(16)30828-1/sref20http://refhub.elsevier.com/S0143-6228(16)30828-1/sref20http://refhub.elsevier.com/S0143-6228(16)30828-1/sref21http://refhub.elsevier.com/S0143-6228(16)30828-1/sref21http://refhub.elsevier.com/S0143-6228(16)30828-1/sref22http://refhub.elsevier.com/S0143-6228(16)30828-1/sref22http://refhub.elsevier.com/S0143-6228(16)30828-1/sref23http://refhub.elsevier.com/S0143-6228(16)30828-1/sref23http://refhub.elsevier.com/S0143-6228(16)30828-1/sref23http://refhub.elsevier.com/S0143-6228(16)30828-1/sref23http://refhub.elsevier.com/S0143-6228(16)30828-1/sref24http://refhub.elsevier.com/S0143-6228(16)30828-1/sref24http://refhub.elsevier.com/S0143-6228(16)30828-1/sref25http://refhub.elsevier.com/S0143-6228(16)30828-1/sref25http://refhub.elsevier.com/S0143-6228(16)30828-1/sref25http://refhub.elsevier.com/S0143-6228(16)30828-1/sref26http://refhub.elsevier.com/S0143-6228(16)30828-1/sref26http://refhub.elsevier.com/S0143-6228(16)30828-1/sref26http://refhub.elsevier.com/S0143-6228(16)30828-1/sref27http://refhub.elsevier.com/S0143-6228(16)30828-1/sref27http://refhub.elsevier.com/S0143-6228(16)30828-1/sref27http://refhub.elsevier.com/S0143-6228(16)30828-1/sref28http://refhub.elsevier.com/S0143-6228(16)30828-1/sref28http://refhub.elsevier.com/S0143-6228(16)30828-1/sref28http://refhub.elsevier.com/S0143-6228(16)30828-1/sref29http://refhub.elsevier.com/S0143-6228(16)30828-1/sref29http://refhub.elsevier.com/S0143-6228(16)30828-1/sref29http://refhub.elsevier.com/S0143-6228(16)30828-1/sref29http://refhub.elsevier.com/S0143-6228(16)30828-1/sref30http://refhub.elsevier.com/S0143-6228(16)30828-1/sref30http://refhub.elsevier.com/S0143-6228(16)30828-1/sref30http://refhub.elsevier.com/S0143-6228(16)30828-1/sref30http://refhub.elsevier.com/S0143-6228(16)30828-1/sref77http://refhub.elsevier.com/S0143-6228(16)30828-1/sref77http://refhub.elsevier.com/S0143-6228(16)30828-1/sref77http://refhub.elsevier.com/S0143-6228(16)30828-1/sref77http://refhub.elsevier.com/S0143-6228(16)30828-1/sref31http://refhub.elsevier.com/S0143-6228(16)30828-1/sref31http://refhub.elsevier.com/S0143-6228(16)30828-1/sref31http://refhub.elsevier.com/S0143-6228(16)30828-1/sref32http://refhub.elsevier.com/S0143-6228(16)30828-1/sref32http://refhub.elsevier.com/S0143-6228(16)30828-1/sref32http://refhub.elsevier.com/S0143-6228(16)30828-1/sref33http://refhub.elsevier.com/S0143-6228(16)30828-1/sref33http://refhub.elsevier.com/S0143-6228(16)30828-1/sref34http://refhub.elsevier.com/S0143-6228(16)30828-1/sref34http://refhub.elsevier.com/S0143-6228(16)30828-1/sref34http://refhub.elsevier.com/S0143-6228(16)30828-1/sref35http://refhub.elsevier.com/S0143-6228(16)30828-1/sref35http://refhub.elsevier.com/S0143-6228(16)30828-1/sref36http://refhub.elsevier.com/S0143-6228(16)30828-1/sref36http://refhub.elsevier.com/S0143-6228(16)30828-1/sref36http://refhub.elsevier.com/S0143-6228(16)30828-1/sref37http://refhub.elsevier.com/S0143-6228(16)30828-1/sref37http://refhub.elsevier.com/S0143-6228(16)30828-1/sref37http://refhub.elsevier.com/S0143-6228(16)30828-1/sref37http://refhub.elsevier.com/S0143-6228(16)30828-1/sref38http://refhub.elsevier.com/S0143-6228(16)30828-1/sref38http://refhub.elsevier.com/S0143-6228(16)30828-1/sref38http://refhub.elsevier.com/S0143-6228(16)30828-1/sref39http://refhub.elsevier.com/S0143-6228(16)30828-1/sref39http://refhub.elsevier.com/S0143-6228(16)30828-1/sref39http://refhub.elsevier.com/S0143-6228(16)30828-1/sref40http://refhub.elsevier.com/S0143-6228(16)30828-1/sref40http://refhub.elsevier.com/S0143-6228(16)30828-1/sref40http://refhub.elsevier.com/S0143-6228(16)30828-1/sref41http://refhub.elsevier.com/S0143-6228(16)30828-1/sref41http://refhub.elsevier.com/S0143-6228(16)30828-1/sref41http://refhub.elsevier.com/S0143-6228(16)30828-1/sref42http://refhub.elsevier.com/S0143-6228(16)30828-1/sref42http://refhub.elsevier.com/S0143-6228(16)30828-1/sref43http://refhub.elsevier.com/S0143-6228(16)30828-1/sref43http://refhub.elsevier.com/S0143-6228(16)30828-1/sref43http://refhub.elsevier.com/S0143-6228(16)30828-1/sref44http://refhub.elsevier.com/S0143-6228(16)30828-1/sref44http://refhub.elsevier.com/S0143-6228(16)30828-1/sref45http://refhub.elsevier.com/S0143-6228(16)30828-1/sref45http://refhub.elsevier.com/S0143-6228(16)30828-1/sref45http://refhub.elsevier.com/S0143-6228(16)30828-1/sref45http://refhub.elsevier.com/S0143-6228(16)30828-1/sref46http://refhub.elsevier.com/S0143-6228(16)30828-1/sref46http://refhub.elsevier.com/S0143-6228(16)30828-1/sref46http://dx.doi.org/10.1007/s12061-015-9165-zhttp://refhub.elsevier.com/S0143-6228(16)30828-1/sref48http://refhub.elsevier.com/S0143-6228(16)30828-1/sref48http://refhub.elsevier.com/S0143-6228(16)30828-1/sref48http://refhub.elsevier.com/S0143-6228(16)30828-1/sref48http://refhub.elsevier.com/S0143-6228(16)30828-1/sref49http://refhub.elsevier.com/S0143-6228(16)30828-1/sref49http://refhub.elsevier.com/S0143-6228(16)30828-1/sref49http://refhub.elsevier.com/S0143-6228(16)30828-1/sref50http://refhub.elsevier.com/S0143-6228(16)30828-1/sref51http://refhub.elsevier.com/S0143-6228(16)30828-1/sref51http://refhub.elsevier.com/S0143-6228(16)30828-1/sref51http://dx.doi.org/10.1177/0042098016664299http://refhub.elsevier.com/S0143-6228(16)30828-1/sref53http://refhub.elsevier.com/S0143-6228(16)30828-1/sref53http://refhub.elsevier.com/S0143-6228(16)30828-1/sref53http://www.citylab.com/housing/2015/09/mapping-the-urban-fingerprints-of-cities/404923/http://www.citylab.com/housing/2015/09/mapping-the-urban-fingerprints-of-cities/404923/http://refhub.elsevier.com/S0143-6228(16)30828-1/sref55http://refhub.elsevier.com/S0143-6228(16)30828-1/sref55http://refhub.elsevier.com/S0143-6228(16)30828-1/sref55http://refhub.elsevier.com/S0143-6228(16)30828-1/sref55

J.D. Barnum et al. / Applied Geography 79 (2017) 203e211 211

Rengert, G. F., Ratcliffe, J. H., & Chakravorty, S. (2005). Policing illegal drug markets: Geographic approaches to crime reduction. Monsey, NY: Criminal Justice Press.

Roncek, D. W., & Bell, R. (1981). Bars, blocks, and crimes. Journal of Environmental Systems, 11(1), 35e47.

Roncek, D. W., & Faggiani, D. (1985). High schools and crime: A replication. The Sociological Quarterly, 26, 491e505.

Roncek, D. W., & Lobosco, A. (1983). The effect of high schools on crime in their neighborhoods. Social Science Quarterly, 64(3), 598e613.

Roncek, D. W., & Maier, P. A. (1991). Bars, blocks, and crimes revisited: Linking the theory of routine activities to the empiricism of “hot spots.” Criminology, 29, 725e753.

Rosenfeld, R., Deckard, M. J., & Blackburn, E. (2014). The effects of directed patrol and self-initiated enforcement on firearm violence: A randomized controlled study of hot spot policing. Criminology, 52(3), 428e449.

Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94(9), 774e802.

Sampson, R. J., Morenoff, J. D., & Gannon-Rowley, T. (2002). Assessing “neighbor- hood effects”: Social processes and new directions in research. Annual Review of Sociology, 28, 443e478.

Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency in urban areas. Chicago, IL: University of Chicago Press.

Sherman, L. W. (1995). Hot spots of crime and criminal careers of places. In J. E. Eck, & D. Weisburd (Eds.), Crime and place (pp. 13e52). Monsey, NY: Criminal Justice Press.

Sherman, L. W., Gartin, P. R., & Buerger, M. E. (1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27, 27e56.

Sherman, L. W., & Weisburd, D. (1995). General deterrent effects of police patrol in

crime “hot spots”: A randomized, controlled trial. Justice Quarterly, 12(4), 625e648.

Skogan, W., & Frydl, K. (2004). Fairness and effectiveness in policing: The evidence. Committee to Review Research on Police Policy and Practices. Committee on Law and Justice, Division of Behavioral and Social Sciences and Education. Washington, D.C: The National Academies Press.

Smith, W. R., Frazee, S. G., & Davison, E. L. (2000). Furthering the integration of routine activity and social disorganization theories: Small units of analysis and the study of street robbery as a diffusion process. Criminology, 38, 489e524.

Spelman, W. (1993). Abandoned buildings: Magnets for crime? Journal of Criminal Justice, 21, 481e495.

St Jean, P. K. B. (2007). Pockets of crime: Broken windows, collective efficacy, and the criminal point of view. Chicago, IL: University of Chicago Press.

Weisburd, D. (2008). Place-based policing. Ideas in American policing. Washington, DC: Police Foundation.

Weisburd, D., Bernasco, W., & Bruinsma, G. (2008). Putting crime in its place: Units of analysis in geographic criminology. New York, NY: Springer.

Weisburd, D., & Green, L. (1995). Policing drug hot spots: The Jersey City drug market analysis experiment. Justice Quarterly, 12, 711e735.

Weisburd, D. L., Groff, E. R., & Yang, S. M. (2012). The criminology of place: Street segments and our understanding of the crime problem. New York, NY: Oxford University Press.

Wortley, R., & Mazerolle, L. (2008). Environmental criminology and crime analysis. New York, NY: Routledge.

Wright, R. T., & Decker, S. H. (1997). Armed robbers in action: Stickups and street culture. Lebanon, NH: Northeastern University Press.http://refhub.elsevier.com/S0143-6228(16)30828-1/sref78http://refhub.elsevier.com/S0143-6228(16)30828-1/sref78http://refhub.elsevier.com/S0143-6228(16)30828-1/sref56http://refhub.elsevier.com/S0143-6228(16)30828-1/sref56http://refhub.elsevier.com/S0143-6228(16)30828-1/sref56http://refhub.elsevier.com/S0143-6228(16)30828-1/sref57http://refhub.elsevier.com/S0143-6228(16)30828-1/sref57http://refhub.elsevier.com/S0143-6228(16)30828-1/sref57http://refhub.elsevier.com/S0143-6228(16)30828-1/sref58http://refhub.elsevier.com/S0143-6228(16)30828-1/sref58http://refhub.elsevier.com/S0143-6228(16)30828-1/sref58http://refhub.elsevier.com/S0143-6228(16)30828-1/sref59http://refhub.elsevier.com/S0143-6228(16)30828-1/sref59http://refhub.elsevier.com/S0143-6228(16)30828-1/sref59http://refhub.elsevier.com/S0143-6228(16)30828-1/sref59http://refhub.elsevier.com/S0143-6228(16)30828-1/sref60http://refhub.elsevier.com/S0143-6228(16)30828-1/sref60http://refhub.elsevier.com/S0143-6228(16)30828-1/sref60http://refhub.elsevier.com/S0143-6228(16)30828-1/sref60http://refhub.elsevier.com/S0143-6228(16)30828-1/sref61http://refhub.elsevier.com/S0143-6228(16)30828-1/sref61http://refhub.elsevier.com/S0143-6228(16)30828-1/sref61http://refhub.elsevier.com/S0143-6228(16)30828-1/sref62http://refhub.elsevier.com/S0143-6228(16)30828-1/sref62http://refhub.elsevier.com/S0143-6228(16)30828-1/sref62http://refhub.elsevier.com/S0143-6228(16)30828-1/sref62http://refhub.elsevier.com/S0143-6228(16)30828-1/sref63http://refhub.elsevier.com/S0143-6228(16)30828-1/sref63http://refhub.elsevier.com/S0143-6228(16)30828-1/sref64http://refhub.elsevier.com/S0143-6228(16)30828-1/sref64http://refhub.elsevier.com/S0143-6228(16)30828-1/sref64http://refhub.elsevier.com/S0143-6228(16)30828-1/sref64http://refhub.elsevier.com/S0143-6228(16)30828-1/sref65http://refhub.elsevier.com/S0143-6228(16)30828-1/sref65http://refhub.elsevier.com/S0143-6228(16)30828-1/sref65http://refhub.elsevier.com/S0143-6228(16)30828-1/sref66http://refhub.elsevier.com/S0143-6228(16)30828-1/sref66http://refhub.elsevier.com/S0143-6228(16)30828-1/sref66http://refhub.elsevier.com/S0143-6228(16)30828-1/sref66http://refhub.elsevier.com/S0143-6228(16)30828-1/sref67http://refhub.elsevier.com/S0143-6228(16)30828-1/sref67http://refhub.elsevier.com/S0143-6228(16)30828-1/sref67http://refhub.elsevier.com/S0143-6228(16)30828-1/sref67http://refhub.elsevier.com/S0143-6228(16)30828-1/sref68http://refhub.elsevier.com/S0143-6228(16)30828-1/sref68http://refhub.elsevier.com/S0143-6228(16)30828-1/sref68http://refhub.elsevier.com/S0143-6228(16)30828-1/sref68http://refhub.elsevier.com/S0143-6228(16)30828-1/sref69http://refhub.elsevier.com/S0143-6228(16)30828-1/sref69http://refhub.elsevier.com/S0143-6228(16)30828-1/sref69http://refhub.elsevier.com/S0143-6228(16)30828-1/sref70http://refhub.elsevier.com/S0143-6228(16)30828-1/sref70http://refhub.elsevier.com/S0143-6228(16)30828-1/sref71http://refhub.elsevier.com/S0143-6228(16)30828-1/sref71http://refhub.elsevier.com/S0143-6228(16)30828-1/sref72http://refhub.elsevier.com/S0143-6228(16)30828-1/sref72http://refhub.elsevier.com/S0143-6228(16)30828-1/sref79http://refhub.elsevier.com/S0143-6228(16)30828-1/sref79http://refhub.elsevier.com/S0143-6228(16)30828-1/sref79http://refhub.elsevier.com/S0143-6228(16)30828-1/sref73http://refhub.elsevier.com/S0143-6228(16)30828-1/sref73http://refhub.elsevier.com/S0143-6228(16)30828-1/sref73http://refhub.elsevier.com/S0143-6228(16)30828-1/sref74http://refhub.elsevier.com/S0143-6228(16)30828-1/sref74http://refhub.elsevier.com/S0143-6228(16)30828-1/sref75http://refhub.elsevier.com/S0143-6228(16)30828-1/sref75

  • The crime kaleidoscope: A cross-jurisdictional analysis of place features and crime in three urban environments
    • 1. Introduction
    • 2. Conceptual framework
    • 3. The study
      • 3.1. Risk terrain modeling
      • 3.2. Data and study settings
      • 3.3. Outcome event
      • 3.4. Potential risk factors
      • 3.5. Model parameters and operationalization
      • 3.6. Statistical analysis
    • 4. Results
    • 5. Discussion
    • Funding
    • References

RESEARCH ARTICLE

Predictive crime mapping

J. Fitterera, T.A. Nelsona* and F. Nathoob

aSpatial Pattern Analysis & Research (SPAR) Laboratory, Department of Geography, University of Victoria, Victoria, BC, Canada; bDepartment of Mathematics & Statistics, University of Victoria, Victoria, BC, Canada

Geographic Information Systems (GIS) have emerged as a key tool in intelligence-led policing and spatial predictions of crime are being used by many police services to reduce crime. Break and entries (BNEs) are one of the most patterned and predictable crime types, and may be particularly amendable to predictive crime mapping. A pilot project was conducted to spatially predict BNEs and property crime in Vancouver, Canada. Using detailed data collected by the Vancouver Police Department on where and when observed crimes occur, the statistical model was able to predict future BNEs for residential and commercial locations. Ideally implemented within a mobile GIS, the automated model provides continually updated predictive maps and may assist patrol units in self-deployment decisions. Future research is required to overcome computational and statistical limitations, and to preform model validation.

Keywords: break and entries (BNEs); predictive mapping; Geographic Information Systems (GIS); intelligence led policing; Vancouver; statistical modeling

Introduction

Intelligence-led policing is a growing discipline where data, analysis, and criminal the- ory are used to guide police allocation and decision-making (Ratcliffe, 2012). Given that crime occurs within a geographical context that includes both space and time, informa- tion to support intelligence-led policing is increasingly map-based and can benefit from platforms that allow integration with Geographic Information Systems (GIS) (Chainey & Ratcliffe, 2005). Developments in mobile GIS technology (Tsou & Kim, 2010) are providing new opportunities for spatially explicit approaches to intelligence-led policing. For example, mapping of mobile phone calls and cell towers has enabled mobile GIS to be used to track perpetrators (Saravanan, Thayyil, & Narayanan, 2013). Another exam- ple of mobile GIS systems used for policing is mobile mapping systems installed in police vehicles that are providing patrol units with near real-time information on crime patterns (Wang, 2012).

In some regions, spatial predictions of crime are already being used by police to reduce crime. For example, the Los Angeles Police Department has used spatial predic- tions of crime to preemptively allocate patrol units and have estimated that geographical criminal intelligence have decreased violent crimes by 5.4% and homicides by 22.6% (Uchida et al., 2012). Similarly, identifying and policing crime hot spots has signifi- cantly reduced calls for service throughout the regions of Minneapolis (Sherman & Weisburd, 1995), Jersey City (Braga et al., 1999; Weisburd & Green, 1995), and Kansas

*Corresponding author. Email: trisalyn@uvic.ca

© 2014 Taylor & Francis

Police Practice and Research, 2015 Vol. 16, No. 2, 121–135, http://dx.doi.org/10.1080/15614263.2014.972618mailto:trisalyn@uvic.cahttp://dx.doi.org/10.1080/15614263.2014.972618

City (Sherman et al., 1995) for a variety of offences including drug, violent, and property crimes.

Break and entries (BNEs) are one of the most patterned and predictable crime types (Short, Brantingham, Bertozzi, & Tita, 2010; Short, D’Orsogna, Brantingham, & Tita, 2009). For example, it has been observed that the probability of a repeat offence increases for the subsequent weeks and for the homes near the original BNE site (Short et al., 2009, 2010). Knowledge acquired during the original BNE regarding possessions, entry, and escape, make nearby homes ‘easy’ targets (Short et al., 2009; Wright & Decker, 1994) that often lead to localized hot spots of property crimes (Farrell, 1995; Johnson et al., 2007; Short & Hestbeck, 1995).

The goal of this paper is to present results from a pilot project to spatially predict property crime in Vancouver, Canada. The model was used to predict crime six times per day, at 4-h intervals, and outputs were designed to integrate with mobile mapping systems that provide maps and statistics of observed crime trends to patrol units. Models were generated using the Vancouver Police Districts (VPD) data on crime as well as additional data on the urban and human environment. To inform the model, we began with exploratory analysis that examined the space–time patterns of BNEs across Vancouver between 2001 and 2012. Based on the results of the exploratory analysis, separate models were developed for commercial and residential properties.

Study area and data

Vancouver is a metropolitan city housing more than 2.3 million residents and 22 neighborhoods (City of Vancouver, 2011) (see Figure 1).

Figure 1. Study area. Vancouver, British Columbia, Canada divided by neighborhood. Neighbour- hood delineations obtained from http://www.vancouver.ca/your-government/open-data-catalogue.aspx.

122 J. Fitterer et al.http://www.vancouver.ca/your-government/open-data-catalogue.aspx

The majority of the population resides in the West End and Downtown proper surrounded by English Bay and Burrard Inlet. Lifestyles span from suburban single-family dwellings to high-rise living in the Downtown, West End, and West Point Grey areas; however all regions, with the exclusion of city parks, are highly urbanized offering many opportunities for break and entry offences (City of Vancouver, 2011). There were over 1.4 million property crimes recorded in the Vancouver metropolitan region between 2002 and 2011 (BC Ministry of Justice, 2012).

Crime data

To predict the spatial and temporal patterns of property crimes, we used point occur- rence residential and commercial BNEs data from 2001 to 2012. These data were pro- vided by the Vancouver Police Department’s PRIME database. Attributes included x, y coordinate locations, reported crime category, and estimated time and date of occur- rence. Crime data were converted to binary grids. To enable map-based prediction of crime probabilities six times per day (1:00–4:00, 5:00–8:00, 9:00–12:00, 13:00–16:00, 17:00–20:00, 21:00–23:00), crime data were formatted as presence/absence grids for six 4-h time intervals using a 200 m by 200 m grid of 3014 cells (see Figure 2).

The transformation resulted in 6,618,744 possible space–time locations for BNE occurrences. A value of one in the grid cell indicated break and entry occurrence, while zero indicated crime absence.

Urban and human environment data

Beyond the pattern of past crime, space–time patterns of future crime were related to the distributions of people, properties, and property types. Several data-sets on the urban

Figure 2. The study area divided into a crime prediction grid with each cell measuring 200 m by 200 m.

Police Practice and Research: An International Journal 123

environment and socio-demographics were integrated into the models. To control for variations in crime associated with variance in population density (Hipp & Roussell, 2013), we used 2011 census dissemination block population estimates (see Figure 3(a)) and LandScan ambient population data (see Figure 3(b), which represent human activity patterns. The urban environment was represented using the national road network of pri- mary and secondary roadways available from the GeoBase data repository (http://www. geobase.ca) (see Figure 3(c)) and street light density (Figure 3(d)), which is a measure of how urban an area is.

The density of a road network can indicate land use type in addition to transporta- tion access for escape and potential sightings of perpetrators (Andresen, 2005). Criminologists have established that property types (Lens, 2013) and value (Pope & Pope, 2012) are key indicators of break and entry risk along with perceived ‘lawless- ness’ of an area (Wilson & Kelling, 1982). To represent social instability, we used point graffiti data collected by municipal workers of the city of Vancouver available on from Vancouver’s open data license (http://vancouver.ca/your-government/open-data-cata logue.aspx) (see Figure 4).

Research-restricted municipal tax information also supplied information on the aver- age property values, number of residential and commercial properties, and the dominant housing types across Vancouver (see Figures 5(a)–(d) and 6). All data were formatted into the same 200 m by 200 m grid format as the crime data.

Figure 3. Vancouver’s demographic and amenity information. Map (a) displays the disaggregated census population count, map (b) the ambient population, and maps (c) and (d) display the road density and streetlight count per 200 m grid.

124 J. Fitterer et al.http://www.geobase.cahttp://www.geobase.cahttp://vancouver.ca/your-government/open-data-catalogue.aspxhttp://vancouver.ca/your-government/open-data-catalogue.aspx

Figure 4. Vancouver’s graffiti rate by 200 m grid cell.

Figure 5. Vancouver’s property information derived from geocoded municipal tax information. Maps (a) and (b) display the number and land values of the residential properties across Vancouver, while maps (c) and (d) display the commercial property counts and values.

Police Practice and Research: An International Journal 125

Methods

Summarizing patterns in observed BNEs

Prior to modeling, we analyzed the space–time dynamics of BNEs. First we conducted frequency graphing of residential and commercial break-ins by hour, day, month, and year from 2001 to 2012 to illuminate trends in criminal behavior. Subsequent to analysis through time, we used density mapping (see O’Sullivan & Unwin, 2010 for method detail) of BNEs from 2001 to 2012 to examine spatial patterns and locate BNE hot spots.

In order to build associations through space and time into the model, we needed to characterize space–time clustering of BNE events. We employed Ratcliffe’s (2009) near- repeat calculator to measure the spatial and temporal distance between each residential and commercial crime event that occurred. We assessed residential and commercial BNEs that were within 500, 850, and 1000 m from the originating event and from one day up to 30 days since the event occurred. Observed patterns were compared to ran- dom patterns to determine if the pattern of repeat or near repeat offences was signifi- cantly different from random.

Predicting future BNEs

Separate predictive models were developed for residential and commercial crimes, as crimes separated by property type had different observed patterns. We also built two models for each property type. Model 1 (Residential and Commercial) was based on integrating crime data and with ancillary data on the urban and human environment (see Table 1), in a regression model (generalized linear logistic regression models with

Figure 6. Vancouver’s dominant property types by 200 m grid cell.

126 J. Fitterer et al.

parameters fit using a logit link function and maximum likelihood estimation) (Burns & Burns, 2008).

Model 2 used only observed crime data in the regression model. Prior to running models, we assessed variable importance (using a Knox ratio statistic) and assessed vari- able correlation to ensure models were statistically robust. Crime data were summarized both for general trends through all years, with more detailed patterns quantified from the past 240 days of data, which maximized detail while keeping computations manage- able. Predictions were mapped and visually compared between observed and predicted spatial patterns to assess model accuracy.

Results

Summarizing patterns in observed BNEs

Exploratory analysis of observed data revealed distinct patterns in the characteristics of break and entry offences across Vancouver. Observing annual trends, we found a

Table 1. The human and urban environmental covariates used in Model 1.

Data Source Description

Road Density

GeoBase National Road Network Version 2 (http://www.geobase.ca)

A collection of primary and secondary roadways at 1:100,000 scale resolution

Dominant property type

Vancouver City/Municipal Tax Information

Dominant property types per 200 m were calculated using a majority and geocoded property locations

Residential property count

Vancouver City/Municipal Tax Information

Property counts were calculated by selecting comprehensive, single and multiple family dwellings, then counting the frequency using a zonal statistic

Residential land value

Vancouver City/Municipal Tax Information

Property values were calculated by selecting comprehensive, single and multiple family dwellings, then averaging the land values using a zonal statistic

Commercial property count

Vancouver City/Municipal Tax Information

Property counts were calculated by selecting commercial dwellings, then counting the frequency using a zonal statistic

Commercial land value

Vancouver City/Municipal Tax Information

Commercial property values were calculated by selecting commercial dwellings, then averaging the land values using a zonal statistic

Number of streetlights

City of Vancouver Open Data Catalogue (http://data.vancouver.ca/ datacatalogue/index.htm)

Coordinate locations of each pole across the city of Vancouver. Street poles were counted by 200 m grid cell

Ambient Population

LandScan (http://web.ornl.gov/sci/ landscan/)

At 1 km resolution, created by disaggregated census population counts across spatial areas using spatial imagery and environmental characteristics to distinguish likely locations of population activity over 24 h

Graffiti rate per 1000 persons

City of Vancouver Open Data Catalogue (see above)

Weekly updated location of graffiti. We counted occurrence rate per 200 m grid and 1000 persons as a surrogate of socioeconomic status

Census population

2011 Canadian Household Census Dissemination block counts were equally proportioned into the 200 m grid

Police Practice and Research: An International Journal 127http://www.geobase.cahttp://data.vancouver.ca/datacatalogue/index.htmhttp://data.vancouver.ca/datacatalogue/index.htmhttp://web.ornl.gov/sci/landscan/http://web.ornl.gov/sci/landscan/

progressive decrease in the frequency of BNEs for both the residential and the commercial properties (see Figures 7(a)–(d) and 8(a)–(d)).

Residential BNEs decreased from 6224 recorded offences in 2001 to 3336 offences in 2012. Similarly, commercial offences decreased from 2535 offences in 2001 to 1692 offences in 2012. For both the residential and the commercial BNEs, there were no sub- stantive month-to-month variations in the frequency of offences (see Figures 7(a)–(d) and 8(a)–(d)). There was a slight increase in the frequency of BNEs on Fridays for both residential and commercial break-ins though the proportional increase only rose by 1–2% over the other weekdays from 2001 to 2012. Contrasting the monthly and week- day temporal trends, we observed distinct differences in the occurrences of residential and commercial BNEs by hour. Residential offences substantially decreased between 1:00 and 6:00 when residents were most likely at home. Conversely, there were three peaks in the frequency of BNEs at 8:00, 12:00, and 18:00. All other times had a moder- ate frequency of occurrence (see Figure 7(a)–(d)). Commercial BNEs occurred most often during the 3:00, 4:00, 5:00, 17:00, and 18:00 intervals, while offences dramati- cally decreased during the daylight hours between 6:00 and 16:00. All other times exhibited a moderate frequency of offences (see Figure 8(a)–(d)).

Spatially, there was consistency in BNE locations from 2001 to 2012 (see Figure 9(a)–(d)).

The downtown northern portion of Vancouver had the majority of BNEs with sections of the West End, and Downtown surrounding neighborhoods of Strathcona, Kitsilano, Fairview, Mount Pleasant, and southcentral regions of Oakridge and Marpole suffering the highest intensity of property crime. Commercial property crimes exhibited linear spatial trends clustering around the 12th Avenue, Knight, and Kingsway highways (see Figure 9(b) and (d)).

Hot spots of commercial BNEs resided in the downtown neighborhood and sections of Fairview and Mount Pleasant neighborhoods that border downtown. There were indi- cations of local hot spots in the southern Arbutus Ridge and Marpole areas; although

Figure 7. 2001–2012 temporal trends of Vancouver’s residential BNEs.

128 J. Fitterer et al.

Figure 8. 2001–2012 temporal trends of Vancouver’s commercial BNEs.

Figure 9. Vancouver’s 2001–2011 break and entry patterns. Maps (a) and (b) depict the frequency of break-ins per 200 m grid and maps (c) and (d) display the smoothed intensity of break-ins from 2001 to 2011 for the residential and commercial properties, respectively.

Police Practice and Research: An International Journal 129

the southern portion of Vancouver had fewer commercial break-ins likely because of the spatial arrangement of commercial properties (see Figure 5(b)).

Predicting future BNEs

Preliminary analysis indicated that residential break-ins had an increased likelihood of repeat and near repeat occurrences up to an 850 m radius and within hours to days of the last event. Within zero to one day after the initial occurrence, there was a 53% chance of reoccurrence up to 850 m from the originating event. Increased probabilities decayed rapidly with time. There was an increased probability at the seventh day after the originating event with the odds of a repeat offence within 850 m increasing by 24%. Commercial BNEs were closely related in time and space, and were influenced within one and two days after the original crime and 500 m away, with a 53 and 21% chance of increasing reoccurrence, respectively.

For Model 1 (residential), 11 data-sets were statistically significant predictors (α = .05). The strongest predictor variables were the count of BNEs up within 850 m from the event in the last 24 h, 24–48 h, and seventh day. Proportion of historical crime by time and day was also significant, as was road density, dominant property type, count of residential crimes in each cell, ambient population, and residential property count. With Model 2 (residential), which was built on only crime data, the same crime variables were significant as in Model 1 (residential).

Model 1 (commercial), which was fit with covariate and crime data, had 14 statisti- cally significant predictor variables (α = .05). These included the count of BNEs up to 500 m away from the crime in the last 24 h and 48 h. Again, proportion of historical crime by time and day were strong predictors. Other reliable variables for prediction included road density, count of commercial crimes from 2001 to 2011 and commercial property count, density of commercial offences from 2001 to 2011, ambient population count, graffiti rate, census population count, commercial property value, and dominant property type. Consistent with Model 2 (residential), with the commercial version of Model 2, we observed that the same crime variables (crime in the last 24 h and 48 h) were significant. We have mapped relative probabilities of possible BNEs based on out- put of Models 1 and 2 (see Figure 10). These maps can be overlaid with observed data BNEs to evaluate the models.

Discussion

Our results contribute to the growing body of literature studying patterns in criminal offences. Following the results of Short et al. (2010) and Johnson et al. (2007), we found both residential and commercial crimes had a strong spatial clustering over short time periods suggesting a near-repeat offence dynamic, and over a longer time frame a core of break and entry offences (Sagovsky & Johnson, 2007; Townsley, Homel, & Chaseling, 2003). Our results indicated that perpetrators prefer to reoffend where they have local knowledge about residents’ routine activities, possessions, and can confirm successful property entry (Wright & Decker, 1994). Recurrent BNEs, in both commer- cial and residential properties, were most likely to occur in the downtown and surround- ing neighborhoods of Vancouver, which is likely the result of a greater population and property base.

Researchers have discussed the need to build crime forecasting capabilities that can be frequently updated (Haberman & Ratcliffe, 2012). Our results indicate that within

130 J. Fitterer et al.

short periods of time and nearby distances of a break-in, there is a significant increase in the chance of another break-in. Therefore, we programmed an automated calculation of crime in the past hours or days from mapped data, available from most police dis- patch systems, and in doing so have increased the functionality and speed of computa- tion for identifying possible future crime locations. Further, by incorporating the most recent counts of the break-ins in the immediate locations with data on historical BNE patterns within the flexible modeling framework, we have provided a functioning and automated approach to updating prediction of break and entry odds every 4 h.

Overall, the predictive model was limited by the relatively rare occurrence, statisti- cally speaking, of both residential and commercial BNEs. Considering that the police recorded 3336 residential break-ins and 1692 commercial break-ins in 2012, the odds of a break-in per grid cell were extremely low, .0005–.0002%, respectively, if every cell and time-interval was considered. The rarity of both the residential and commercial break-in compared to the 4-h modeling interval posed some modeling limitations. These limitations were largely a function of the temporal scale. Over a month to a year, there is a clear indication of break and entry clustering; however, when assessed multiple times per day, the pattern is more difficult to predict. For any crime event, the environ- mental characteristics and count of crimes in the last hours or days in adjacent cells may be identical, and in one circumstance a repeat break and entry occurs and in the other it does not. This ‘randomness’ leads to a reduced strength of the explanatory power of the covariates, especially in the case of the residential break-in model where BNEs have occurred in almost every grid cell over 2001–2011.

Studies have demonstrated that crime prevention strategies targeted to hot spots have led to substantial reductions in offences (Braga, 2001, 2005; Braga, Kennedy, & Bond, 2008; Weisburd et al., 2006). Models that incorporate both people and place characteris- tics, as well as crime data, may be suitable for mapping relative risk of BNEs. However, our preliminary investigations suggest that false positives, or indications of break and

Figure 10. Example of a comparison of observed break-ins compared to probability of a break-in.

Police Practice and Research: An International Journal 131

entry likelihood where none will occur, may be high. The emphasis of these models on the nature of the environment may be more appropriate for predicting crimes over broader time periods, such as monthly trends. Conversely, models built largely from the crimes occurring in the last 24 and 24–48 h pinpointed intersections of crime probability based on the spatial proximity to the last break and entry. These results can be used as flags of potential break and entry in subsequent hours and day. However, it can be diffi- cult to predict these at fine spatial resolutions given the importance of spatial neighbor- hoods of 850 m for residential and 500 m for commercial crimes.

A unique aspect of our research is that the model we developed is statistical and based entirely on data. Other similar models have used mathematical approaches such as random walks (Jones, Brantingham, & Chayes, 2010; Short et al., 2009) to represent processes of crime. Given the fine detail of VPD’s data and long-time series of available data, a statistical model can harness past information on crime to predict future patterns. However, our research suggests that using only crime data may limit the spatial and temporal resolution of the model. For instance, when only using the crime data, it will be difficult to generate a model at a finer spatial scale than 850 m for residential proper- ties and 500 m for commercial properties. These areas are too large to be feasible for tasking police patrol units. By including data on the urban and human context, the finer grid cells model is useful, but initial investigations indicate the temporal pattern of crime predicted may be more appropriate for representing monthly patterns.

Our model predicted BNEs six times per day. The time periods were limited by the size of the data-sets, the accuracy of event time recorded in the crime data, and the fre- quency with which a system for implementation would update data. When structured for a model that predicted six times per day, the observed data were stored in over 6 million cells. One hour models would require close to 150 million cells, making compu- tational limitations a serious problem. As well, the success of the model would be lim- ited by variation in when the crime occurred relative to when the crime was reported, which is difficult to correct to within an hour time frame. In an operational context, the frequency of predictive mapping, which is based on the temporal detail of data, will always be limited to the frequency in which new data are added to the system. As such, our recommendation is that future researchers should consider approaches that integrate statistically or data-based methods with mathematical approaches for modeling known processes of crime. Harnessing the power of growing data-sets with knowledge of crimi- nal patterns may allow us to customize model parameterization, while mathematical approaches will enable more detailed space and time predictions.

The maps provided are appropriate for relative risk assessment of the likelihood of BNEs (e.g. Caplan, Kennedy, & Miller, 2011). In order to assess how well the models are predicting crime, a method for converting likelihood into expected crime is needed. The standard approach would be to threshold the likelihood surface and above a thresh- old to classify the crime as predicted. The predicted crime can then be compared to observed testing data to determine the accuracy and number of false positives. Given the low probability of any crime occurring, it is unclear how to set the threshold for identifying a predicted crime event. If the value is too high, there will be an overwhelm- ing number of false predictions. However, setting the threshold too low may lead to the model missing most of the real crime. There may be benefit to considering spatial or temporal pattern information in setting the threshold. For instance, given that all crime is unlikely, it may be when there is a large change in the predicted value for a specific grid cell that the potential for crime should be highlighted. Further research is required before the model can be fully assessed or implemented.

132 J. Fitterer et al.

An important component of implementation is the automation of the predictive system. Modeling BNE risk multiple times per day is only feasible if it can be per- formed without analyst intervention, particularly when predictions are required outside typical working hours. Ideally, this model will be automated through integration within mobile GIS-style systems. For instance, systems are being developed to outfit police vehicles with GIS mapping that provides insight into historical trends and recent changes in activity. Such systems are the ideal platform for integrating automated predictive mapping capabilities that are designed for allocating patrol resources.

Conclusion

Intelligence-led policing is supported by spatially explicit models for predicting where and when future crimes will occur. Our models predicted future BNEs for residential and commercial locations based on detailed data collected by the VPD on where and when observed crimes occur. Models predicted BNE locations six times per day and the models are designed to be implemented in a mobile GIS system that will allow for auto- mated updating. We have generated models based on both crime and urban and human environmental data, as well as on only crime data. When models considered the envi- ronmental context, the predictions had great spatial detail (200 m by 200 m), but the patterns seem to be most representative of long-term trends. When only crime data were used in predictions, the patterns were representative of short time periods, but the spatial detail of maps is lower (500–850 m). For patrol purposes, finer spatial and temporal resolutions are required. The next phase of models should integrate statistical data-based models, such as we have presented here, with mechanistic models that base predictions on mathematically represented knowledge of typical crime patterns. An integrated modeling approach will allow the powerful crime data to be leveraged in a dynamic framework. As well, the model needs to be further evaluated by converting crime likeli- hood maps into maps of predicted crime and comparing the model results with observed data. Model validation will require thoughtful consideration of how to threshold likelihood surfaces to predict crime events.

Acknowledgments This work was support by NSERC Engage. Thanks to Francis Graf of Latitude Geographic Inc. and Ryan Prox of the Vancouver Police Division who provided direction, data, and input for this work. We would like to thank Susan Kinniburgh from the University of Victoria for help with model coding.

Notes on contributors J. Fitterer is a PhD student in the SPAR Lab. She is currently applying her expertise in spatial data analysis to research questions on how spatial and temporal patterns of alcohol access impact violent crime.

Dr T.A. Nelson is director of SPAR Lab and a Research Chair Spatial Sciences in the Department of Geography at the University of Victoria. She brings expertise in spatial–temporal analysis and modeling to a wide range of research applications including ecology and public health.

Dr F. Nathoo is an Associate Professor and Canada Research Chair in the Department of Mathe- matics and Statistics the University of Victoria. He is an expert in spatial Bayesian statistics and high-dimensional data. He focuses on large-scale problems and has as developed innovative statistical and computing methods for a range of applications.

Police Practice and Research: An International Journal 133

References Andresen, M. A. (2005). Crime measures and the spatial analysis of criminal activity. British

Journal of Criminology, 46, 258–285. doi:10.1093/bjc/azi054 BC Ministry of Justice. (2012). Regional profile 2011: Metro Vancouver Region (pp. 1–11).

Retrieved from http://www.pssg.gov.bc.ca/policeservices/statistics/docs/14_Metro_Vancouver. pdf

Braga, A. A. (2001). The effects of hot spots policing on crime. The Annals of the American Academy of Political and Social Science, 578, 104–125. doi:10.1177/000271620157800107

Braga, A. A. (2005). Hot spots policing and crime prevention: A systematic review of randomized controlled trials. Journal of Experimental Criminology, 1, 317–342. doi:10.1007/s11292-005- 8133-z

Braga, A. A., Kennedy, J. F., & Bond, B. J. (2008). Policing crime and disorder hot spots: A ran- domized controlled trial. Criminology, 46(3), 1–31.

Braga, A. A., Weisburd, D. L., Waring, E. J., Mazerolle, L. G., Spelman, W., & Gajewski, F. (1999). Problem-oriented policing in violent crime places: A randomized controlled experi- ment. Criminology, 37, 541–580. doi:10.1111/j.1745-9125.1999.tb00496.x

Burns, R., & R. Burns. (2008). Business research methods and statistics using SPSS (33(1) ed., p. 544). Portland, OR: Book News.

Caplan, J. M., Kennedy, L. W., & Miller, J. (2011). Risk terrain modeling: Brokering criminologi- cal theory and GIS methods for crime forecasting. Justice Quarterly, 28, 360–381.

Chainey, S., & Ratcliffe, J. (2005). GIS and crime mapping. West Sussex: Wiley. City of Vancouver. (2011). Census local area profiles 2011. Retrieved from http://data.vancouver.

ca/datacatalogue/censusLocalAreaProfiles2011.htm Farrell, G. (1995). Preventing repeat victimization. Crime and Justice, 19, 469–534. Haberman, C. P., & Ratcliffe, J. H. (2012). The predictive policing challenges of near repeat

armed street robberies. Policing, 6, 151–166. doi:10.1093/police/pas012 Hipp, J. R., & Roussell, A. (2013). Micro- and macro-environment population and the conse-

quences for crime rates. Social Forces, 92, 563–595. doi:10.1093/sf/sot091 Johnson, S. D., Bernasco, W., Bowers, K. J., Elffers, H., Ratcliffe, J., Rengert, G., & Townsley,

M. (2007). Space–time patterns of risk: A cross national assessment of residential burglary victimization. Journal of Quantitative Criminology, 23, 201–219. doi:10.1007/s10940-007- 9025-3

Jones, P. A., Brantingham, P. J., & Chayes, L. R. (2010). Statistical models of criminal behaviour: The effects of law enforcement actions. Mathematical Models & Methods in Applied Sciences, 20, 1397–1423.

Lens, M. C. (2013). Subsidized housing and crime: Theory, mechanisms, and evidence. Journal of Planning Literature, 28, 352–363. doi:10.1177/0885412213500992

O’Sullivan, D., & Unwin, D. J. (2010). Geographic information analysis (2nd ed., p. 405). Wiley. Retrieved from http://www.knovel.com/web/portal/browse/display?_EXT_KNOVEL_DIS PLAY_bookid=5023

Pope, D. G., & Pope, J. C. (2012). Crime and property values: Evidence from the 1990s crime drop. Regional Science and Urban Economics, 42, 177–188. doi:10.1016/j.regsciurb- eco.2011.08.008

Ratcliffe, J. (2009). Near repeat calculator. Philadelphia, PA: Temple University and National Institute of Justice. Retrieved from http://www.temple.edu/cj/misc/nr/download.htm

Ratcliffe, J. (2012). Intelligence-led policing (p. 288). Portland, OR, Willan Publishing. Sagovsky, A., & Johnson, S. D. (2007). When does repeat burglary victimisation occur?

Australian & New Zealand Journal of Criminology, 40(1), 1–26. doi:10.1375/acri.40.1.1 Saravanan, M., Thayyil, R., & Narayanan, S. (2013). Enabling real time crime intelligence using

mobile GIS and prediction methods. Intelligence and Security Informatics Conference (EISIC), 2013 European. Uppsala: IEEE, 2013.

Sherman, L. W., Rogan, D. P., Edwards, T., Whipple, R., Shreve, D., Witcher, D., … Bridgeforth, C. A. (1995). Deterrent effects of police raids on crack houses: A randomized, controlled experiment. Justice Quarterly, 12, 755–781. doi:10.1080/07418829500096281

Sherman, L. W., & Weisburd, D. (1995). General deterrent effects of police patrol in crime ‘hot spots’: A randomized, controlled trial. Justice Quarterly, 12, 625–648. doi:10.1080/ 07418829500096221

134 J. Fitterer et al.http://dx.doi.org/10.1093/bjc/azi054http://www.pssg.gov.bc.ca/policeservices/statistics/docs/14_Metro_Vancouver.pdfhttp://www.pssg.gov.bc.ca/policeservices/statistics/docs/14_Metro_Vancouver.pdfhttp://dx.doi.org/10.1177/000271620157800107http://dx.doi.org/10.1007/s11292-005-8133-zhttp://dx.doi.org/10.1007/s11292-005-8133-zhttp://dx.doi.org/10.1111/j.1745-9125.1999.tb00496.xhttp://data.vancouver.ca/datacatalogue/censusLocalAreaProfiles2011.htmhttp://data.vancouver.ca/datacatalogue/censusLocalAreaProfiles2011.htmhttp://dx.doi.org/10.1093/police/pas012http://dx.doi.org/10.1093/sf/sot091http://dx.doi.org/10.1007/s10940-007-9025-3http://dx.doi.org/10.1007/s10940-007-9025-3http://dx.doi.org/10.1177/0885412213500992http://www.knovel.com/web/portal/browse/display?_EXT_KNOVEL_DISPLAY_bookid=5023http://www.knovel.com/web/portal/browse/display?_EXT_KNOVEL_DISPLAY_bookid=5023http://dx.doi.org/10.1016/j.regsciurbeco.2011.08.008http://dx.doi.org/10.1016/j.regsciurbeco.2011.08.008http://www.temple.edu/cj/misc/nr/download.htmhttp://dx.doi.org/10.1375/acri.40.1.1http://dx.doi.org/10.1080/07418829500096281http://dx.doi.org/10.1080/07418829500096221http://dx.doi.org/10.1080/07418829500096221

Short, M. B., Brantingham, P. J., Bertozzi, A. L., & Tita, G. E. (2010). Dissipation and displace- ment of hotspots in reaction–diffusion models of crime. Proceedings of the National academy of Sciences of the United States of America, 107, 3961–3965. doi:10.1073/pnas.0910921107

Short, M. B., D’Orsogna, M. R., Brantingham, P. J., & Tita, G. E. (2009). Measuring and model- ing repeat and near-repeat burglary effects. Journal of Quantitative Criminology, 25, 325–339. doi:10.1007/s10940-009-9068-8

Short, H., & Hestbeck, J. B. (1995). National biotic resource inventories and GAP analysis. BioScience, 45, 535–539.

Townsley, M., Homel, R., & Chaseling, J. (2003). Infectious burglaries: A test of the near repeat hypothesis. British Journal of Criminology, 43, 615–633.

Tsou, M.-H., & Kim, I. H. (2010). Increasing spatial awareness by integrating Internet Geographic Information Services (GIServices) with real time wireless mobile GIS applications. Interna- tional Journal of Strategic Information Technology and Applications, 1, 42–54.

Uchida, C., Swatt, M., Gamero, D., Lopez, J. Salazar, E., King, E., … Michael, D. (2012). Los Angeles California smart policing initiative (pp. 1–12). Washington, DC: Bureau of Justice Assistance, US Department of Justice.

Wang, F. (2012). Why police and policing need GIS: An overview. Annals of GIS, 18, 159–171. doi:10.1080/19475683.2012.691900

Weisburd, D., & Green, L. (1995). Policing drug hot spots: The Jersey City drug market analysis experiment. Justice Quarterly, 12, 711–735. doi:10.1080/07418829500096261

Weisburd, D., Wyckoff, L. A., Ready, J., Eck, J. E., Hinkle, J. C., & Gajewski, F. (2006). Does crime just move around the corner? A controlled study of spatial displacement and diffusion of crime control benefits. Criminology, 44, 549–592.

Wilson, J. Q., & Kelling, G. L. (1982). Broken windows. The Atlantic Online, March 1–9. Wright, R., & Decker, S. (1994). Burglars on the job. Boston, MA: Northeastern University

Press.

Police Practice and Research: An International Journal 135http://dx.doi.org/10.1073/pnas.0910921107http://dx.doi.org/10.1007/s10940-009-9068-8http://dx.doi.org/10.1080/19475683.2012.691900http://dx.doi.org/10.1080/07418829500096261

Copyright of Police Practice & Research is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.

  • Abstract
  • Introduction
  • Study area and data
    • Crime data
    • Urban and human environment data
  • Methods
    • Summarizing patterns in observed BNEs
    • Predicting future BNEs
  • Results
    • Summarizing patterns in observed BNEs
    • Predicting future BNEs
  • Discussion
  • Conclusion
  • Acknowledgments
  • Notes on con„trib„u„tors
  • References