Geographic Crime Analysis

RESEARCH ARTICLE

GIS supporting intelligence-led policing

Tegan Herchenradera* and Steven Myhill-Jonesb

aLatitude Geographics, Kitchener, Canada; bLatitude Geographics, Victoria, Canada

Tightening budgets and increased demand for public accountability has placed additional stress on already limited police department resources. Web-based crime mapping provides significant improvement over previous methods of information dissemination, allowing police departments to continue to work quickly and effi- ciently within these limitations. This modern technology has enabled a more proac- tive approach to policing, including intelligence led-policing and public facing crime maps. As such, officers are now able to better consider spatial patterns related to historic crime, and determine more informedly where crimes may occur in the future, and allocate their limited resources accordingly.

Keywords: intelligence-led policing; transparency; GIS; web-based mapping; ArcGIS®; Geocortex®

Introduction

In an information-driven society, police departments are under increasing pressure to run an intelligence-led police model. This model asserts that police can spend less time reactively responding to crime if supported by a system that provides data analysis and crime intelligence, allowing officers to reduce, disrupt, and prevent crime (Ratcliffe, 2008, n.d.). Alongside this drive for information is the ongoing demand for departments to provide increased transparency to the media and citizens. The Waterloo Region Police Service (WRPS) and the Vancouver Police Department (VPD) are two Canadian organi- zations which have taken the use and sharing of information to the next level through the implementation of an intelligence-led policing model. As this paper will explore, this has been supported, in part, by providing web-based mapping and basic geographic data analysis capabilities to an expanded audience of stakeholders. In addition to empowering police officers with the information they need to do their jobs better, this work has been naturally extended to serve transparency goals by simultaneously deliver- ing a subset of these data and application capabilities to the general public.

In both the WRPS and the VPD, Geographic Information System (GIS) technology is viewed as a means by which the organization can work more proactively to analyze and prevent crime. A GIS solution ‘integrates hardware, software, and data for captur- ing, managing, analyzing, and displaying all forms of geographically referenced infor- mation’ (‘What is GIS?’). This allows users ‘to view, understand, question, interpret, and visualize data in many ways that reveal relationships, patterns, and trends in the form of maps … reports, and charts’ (‘What is GIS?’). Both WRPS and VPD have had long-standing enterprise GIS deployments based on ESRI® ArcGIS® technology. Given the movement towards an-intelligence led policing model, they sought to extend the

*Corresponding author. Email: [email protected]

© 2014 Taylor & Francis

Police Practice and Research, 2015 Vol. 16, No. 2, 136–147, http://dx.doi.org/10.1080/15614263.2014.972622

capabilities of existing desktop technology through the development of web-mapping applications with assistance from Latitude Geographics and their Geocortex® software technology for ArcGIS® Server.

With mature GIS implementations already in place, web-based mapping enables organizations to reach a wider audience and more fully leverage their investment in GIS technology by using GIS-publishing platforms like ESRI®’s ArcGIS® Server and ArcGIS® Online. These technologies allow organizations to publish their spatial data and related information to the web in the form of services and applications. The services include base maps which show a basic representation of the geography, as well as layers which are a visual representation of discrete types of features, such as property bound- aries, building footprints, or census data. Geocortex® helps organizations build applica- tions which consume the published services and introduce various visualization and analytical tools which can be used by end users.

Key advantages of using a highly configurable commercial off-the-shelf (COTS) solution like Geocortex® come from the significant amount of pre-built and easily con- figurable functionality that adapts over time as technologies progress, the regular addi- tion of new capabilities and options, and the amortization of development costs across numerous licensee organizations. Alternatively, much of the functionality offered by Geocortex® would need to be developed by in-house developers or through third-party professional services. For example, the mapping viewer (which allows a user to view the maps and layers published through ArcGIS® Server and/or ArcGIS® Online) and associated capabilities might typically be developed as custom code or built using free templates as a starting point. Properly engineered COTS solutions can help public safety organizations deliver applications more quickly and focus on domain-specific business problems instead of financing the one-off development of software applications and infrastructure that invariably require subsequent ongoing investment to keep pace with a rapidly changing technology space.

Following the intelligence-led policing model, WRPS and VPD emphasized making high-quality current data available to officers in their patrol cars to help them be more proactive and informed in their patrol tactics. The opportunity to be more forward-looking in their actions is due to the capacity of empirical data to complement an officer’s experience, hunches, and instincts related to geographic attention and pattern recognition. The applications currently show officers information on crime occurrences across their district for specified time periods. As the applications evolve over time, the plan is to add other types of information to the maps, such as lists of known sex offenders or individuals on parole (Herchenrader, personal communication, 13 August 2013; 6 September 2013).

Fulfilling the initial objectives for increased public transparency has been met through development of public websites that display generalized occurrence information suitable for public consumption and the protection of privacy. Citizens are able to visu- alize crimes across a general area as well as in defined locations (e.g. their neighbor- hood or child’s school).

The goal of this study is to examine the usefulness of web-based GIS and mapping applications in a police setting using two real-world Geocortex®-based implementations as case studies. To do so, we will outline how each of the respective police services dis- seminated information to their officers and to the public prior to the implementation of the Geocortex® solution, what issues both VPD and WRPS experienced with these methods, what the Geocortex® solution entailed, what the challenges were with

Police Practice and Research: An International Journal 137

implementing the solution, how the VPD and WRPS plan on developing the application in the future, and what the feedback has been from both officers and the public.

Waterloo regional police service

The problem

Prior to their Geocortex® implementation, the WRPS informed their officers going out for patrol through two methods: paper briefings and internal message boards. To inform the public about crimes in their neighborhood or in the region in general, the Service posted maps rendered in static PDF format of the jurisdiction on their website. These methods of supplying information to officers and the public had enduring drawbacks that warranted attention.

The internal electronic message board available to officers allowed them to post information regarding an incident that occurred during their patrol. A limitation to this method was the time required for the officer to sit down and write a post. Given various time constraints, their availability to do so was at worst minimal and at best variable. Posting to the board was not mandatory and it was up to officers to make time to write about incidents. As such, the method could not be relied upon to be kept up to date on a consistent basis. Though any entry was helpful, by its nature, it was an incomplete data source that offered limited potential for consistent use or meaningful pattern recog- nition. Another limitation of this method was that there was no way to search the board for particular items. Officers gathered information by scrolling through posts. As such, it was easy for officers to miss information or be unaware of it altogether. Paper brief- ings, created by the Service’s crime analyst, occurred at the beginning of each shift. Briefings could be missed for a variety of reasons, such as illness or rushing out due to a call (Herchenrader, personal communication, 13 August 2013).

Given that the information provided in the briefing was not available afterwards, the Service was experiencing an inefficient use of already time-constrained resources. First, the Service’s crime analysts were regularly being asked routine questions, thus taking their time away from other important tasks. Second, during an officer’s downtime on patrol, they were more likely to place themselves in a location that was ‘convenient and safe’ (Herchenrader, personal communication, 13 August 2013), meaning they would go somewhere which their previous experience informed them would be a likely place for problems to occur. Readily available and up-to-date information could more accurately and precisely inform an officer so they could locate themselves at a particular block or building, or at a new and previously unknown location where crime would be more pos- sible to occur.

To inform the public about incidents in the region, static maps of the region were made available on the Service’s website. While these maps provided a wealth of infor- mation at a defined map scale, this became a drawback in coming to any useful conclu- sions. There were many different symbols on the map indicating different types of crime and due to the inability to zoom into the map, it was difficult for the user to get a proper understanding of what was going on in any particular area.

The solution

In the move towards an intelligence-led policing model, as well as to provide insight and transparency to the public, the WRPS decided that a third-party GIS solution, which

138 T. Herchenrader and S. Myhill-Jones

offered a dynamic, user-adjustable map populated with current information, was the answer. They sought to deliver this through an offering of several interactive mapping applications, with appropriate data, visualization, and analysis tools for each intended audience. Spatially visualizing and highlighting specific crime data types makes it easier for officers to observe and draw correlations between occurrences. Over time, this also helps officers better identify and track crime as it increases or decreases and shifts or maintains its location (Gotway & Schabenberger, 2009). Analysis can also be extended beyond the proximity of the crimes. Geospatial data can also allow officers to take into account variables such as neighborhood type, street accessibility, type of property (Malleson, 2011) as well as various other factors that relate to the ‘multidimensional, multifaceted crime problem’ (Rich, 1995). Being enabled to account for a variety of factors can help officers draw more informed conclusions concerning where crime might occur in the future (Malleson, 2011). The interactive functionality of these mapping systems can also provide information that enables officers to better use available resources to work towards the prevention of crime.

To expand upon their pre-existing ESRI® GIS implementation, the Service opted to go with a third-party solution. Reasons included a lack of specialized resources at the Service that could deliver a product that not only met the needs of the Service but that could also be delivered in the allotted time with the given funds (Herchenrader, personal communication, 23 July 2013; Rich, 1995). After reviewing proposals, web-mapping software firm Latitude Geographics (developers of Geocortex®) was selected. With a mature, established product in the ESRI® COTS space, Geocortex® provided the most required capabilities out of the box. With minimal configuration required to deploy, the IT team was able to get the core capabilities deployed quickly, and turn their attention instead to configuration of domain-specific capabilities and data integration (Herchenrader, personal communication, 23 July 2013).

The three primary features of the internal Geocortex® application include a targeted query tool (see Figure 1), charting of query results (see Figure 2) as well as a time slider (see Figure 3) to view crime occurrences over time.

WRPS also added a warrants layer so that officers could query for and view images of individuals with an outstanding arrest warrant. The targeted query tool allows officers to query precisely across dates (start and end date, days of the week, hours of the day), dispatch codes, and areas of the jurisdiction. Depending on the extent, results from a query appear in clusters, with the number of results indicated at the center of the cluster, making it easier to identify hotspots when the user is zoomed out (see Figure 3). As the user zooms into the map, these clusters disperse, indicating the individual occurrences. When an officer clicks on a cluster of occurrences, the details of the various occurrences appear in the results sidebar. An officer can select one of these occurrences to open up a window which provides further details. Officers can perform multiple queries and have these results appear as different layers on a map, providing a powerful analysis tool to use while on patrol.

The time slider and charting tools are both used in conjunction with the query tool. The charting allows officers to graph occurrence results either by line, bar, or pie chart, providing an alternative analytical representation. Given the power and flexibility of the tool, officers may require some additional training and experience to employ it to its full potential (Herchenrader, personal communication, 23 July 2013). Officers who do not have the skill set or inclination to create these charts may continue to push many data analysis tasks back upon the crime analysts. Nonetheless, it is there for those who, of their own accord, see value in coupling their investigative skills with data exploration.

Police Practice and Research: An International Journal 139

The time slider, on the other hand, allows officers to animate the data and show how they appear over time, rather than simultaneously for the entire specified date range. Officers can define a time range and animate the data over that range or view specific instances showing a particular moment in time within a given subset of data.

The public map allows citizens to query recent occurrences as well as historical occurrences. When querying occurrences, the precise locations of crimes are generalized to approximate locations to avoid data misinterpretation and to protect privacy. When querying for recent occurrences, users can select one or multiple occurrence types as well as the zone or division of interest. Individual occurrences are then mapped, with result details appearing in the results sidebar. Each result in the sidebar shows a date range in which the occurrence happened and provides a link for users to contact the Ser- vice with more information about the incident. When querying historical occurrences, users can also select up to six occurrence types, provide a start and end date of the range they are interested in, as well as select the days of the week that they wish to include. The map then color codes the zones of the region, given the number of occur- rences that happened there. Clicking on a zone lists the details of the query as well as the total number of occurrences per crime type. The public map can be accessed at: http://maps.policereporting.ca/SilverlightMap/Viewer.html?Viewer=OccurrenceMaps. In addition to keeping the public informed and the police accountable, public crime maps have the added benefit of mitigating the public’s fear of victimization, not only by showing where crime occurs, but also by sanitizing or otherwise limiting the level of detail of information that they can access about specific crimes (Ratcliffe, 2002).

As with any project, there were challenges. The largest of these was the short time- line allocated for the project. The Service had been given a grant with a specified end

Figure 1. WRPS-targeted query tool.

140 T. Herchenrader and S. Myhill-Jones

date and the project needed to be completed by this date, regardless of obstacles experi- enced. The project timeline was further shortened by the amount of time that it took for Latitude Geographics and related contractors to get clearance to work with the sensitive police data (Herchenrader, personal communication, 23 July 2013).

Figure 2. WRPS charting tool.

Figure 3. WRPS time slider and result clustering.

Police Practice and Research: An International Journal 141

The second challenge was assembling source data to efficiently feed the web- mapping applications. The first issue was that the existing data came from various sources with varying levels of quality assurance and consistency. This meant that much of the data needed to be massaged to make it consistent and readable by the application (Herchenrader, personal communication, 23 July 2013). The other issue with organizing the data was the amount of information related to an event. Arrests, for example, include person, location, crime type, etc. and each of these can be related to other elements or events. As such, instead of pulling directly from the data, the team needed to build summary tables that would be able to make those connections and present them in a usable way (Herchenrader, personal communication, 12 July 2013).

A third challenge was maintaining the scope of the project. WRPS deals with a wide variety of events and the team was continually pressured to offer, from the outset, other data layers, in addition to occurrence information and warrants. While such additions provide value to many, the intent of the initial implementation was to quickly deploy technology provided by Latitude Geographics to establish a foundation from which the Service could continue to build on, over time (Herchenrader, personal communication, 23 July 2013).

Following the launch, WRPS received significant positive feedback, both from offi- cers and the public. Officers report that they find the application easy to use and have found it useful. It is believed this is partially due to the application infrastructure, but also a result of the considerable data improvement that occurred during the project. For the public website, traditional news media have been the Service’s biggest source of feedback, stating that the application is easy to use and provides the necessary informa- tion they require (Herchenrader, personal communication, 13 August 2013). The public site endeavors to provide maximum possible information, but given the sensitive nature of the information, this amounts to fairly limited detail and locational precision about occurrences. Nonetheless, the Service has been able to reduce staff time consumed by directing inquiries to the public map. The Inspector for media and public affairs noted that where before it may have taken him five to thirty minutes to look up an inquiry, it now takes him seconds (Herchenrader, personal communication, 13 August 2013). Hav- ing occurrence information readily available to officials and the public helps to ‘diffuse potential conflicts between community and police’ (Friedman, Gordon, & Maltz, 2000) as management is able to address inquiries with ‘more timely, focused information’ (Friedman et al., 2000) that allows them to ‘correct inaccurate accounts’ (Friedman et al., 2000), while the general public and media are also able to approach interactions better informed themselves.

The Service has planned for the continued development and enhancement of the internal application. A significant amount of effort will go towards adding more layers to the map, including those for compliance checks and high-risk offenders. As before, a challenge will be defining and maintaining the scope of the project, as well as ensuring that resources continue to be allocated towards its development (Herchenrader, personal communication, 23 July 2013).

Vancouver Police Department

The problem

Unlike WRPS, prior to implementing their Geocortex®-based solution, the VPD had previously hired a consultant to implement a customized ArcGIS® Server-based public safety application, referred to internally as GeoDash. Like the Geocortex® solution for

142 T. Herchenrader and S. Myhill-Jones

WRPS, the GeoDash application allowed VPD staff to use basic analytical tools to iden- tify and explore crime patterns and trends on their own, rather than requiring the ser- vices of a crime analyst. In turn, this allowed the Department’s crime analysts to spend their time doing higher value analysis. The GeoDash application allowed officers to view crime distribution, as well as the ability to add a distance buffer to visualize and measure where and how close crimes are in relation to specific locations, such as schools, transit centers, and so on. A built-in reporting feature also allowed users to view crime by concentration, type, and transition at daily, weekly, or monthly intervals.

Although it was generally regarded as an improvement over previous approaches, which involved data redundancy, duplication of effort, and the manual transfer of data between systems, the first-generation GeoDash application had some important limita- tions due to the original development technology on which it was built. The most sig- nificant limitation was that the application was accessible only while in the office via PCs connected to the web. Officers often had insufficient time to gather the information they needed before going out on patrol. It was therefore difficult for them to collate enough relevant information that would adequately inform their patrol, making the application ineffective for officers out on the beat (Manning, 2001). This rush to get enough information before departure was further complicated by the application’s some- what unwieldy-to-navigate menu structure, to which underlying alterations were chal- lenging, given the generation of the original development technology. The application as it was customized also did not include the level of charting and analysis capabili- ties required by the CompStat accountability standard that the VPD subscribes to (Herchenrader, personal communication, 13 August 2013). For example, the first- generation GeoDash application was not developed to compare time of day and/or day of week crimes by location or buffer, nor did the application ‘establish threshold analysis of specific crime types by CompStat period’ (Herchenrader, personal communication, 13 August 2013).

The solution

Since the primary issue was limited access to required data and analysis capabilities to support patrol activities, the VPD considered developing an additional web-mapping application to provide up-to-date information and crime analysis tools to patrol officers in their vehicles. It was hypothesized that by providing basic crime analysis capabilities to the patrol officer, they could combine bigger picture data with their own instincts and experience to be more proactive in their patrols. Crime analysts would be free to pursue more complex crime analysis, rather than responding to routine inquiries. After weighing the merits of different approaches, the Department decided to again partner with a third-party vendor as they did not have the in-house expertise with available cycles to undertake or maintain a custom software development project of this envisioned magnitude. Latitude Geographics was selected as the vendor, and its Geocortex® software was selected as the COTS solution based on its successful track record being used by other organizations deploying ESRI®’s ArcGIS® Server custom- ers and its demonstrated ability to deliver a solution meeting the specific requirements of the Department (Herchenrader, personal communication, 13 August 2013).

The three primary goals for the Geocortex® solution were: deployment to the touch- screen interface of the Panasonic Toughbook Mobile Display Terminals (MDT) in patrol cars, simplification and enrichment of the user experience, and minimization of data

Police Practice and Research: An International Journal 143

transferred to the mobile application (to reduce cellular data costs) (Herchenrader, personal communication, 23 July 2013).

To address requirements around deploying to the MDTs and a revised user interface required for an in-car, touchscreen environment, the user interface of the mapping viewer was modified. Though Geocortex® provides an out-of-the-box view intended for mobile browsing, it required further refinement to be deployed on the MDT PCs that are oper- ated largely by keyboard and touch-based input (instead of a mouse or related pointer). Given the constraint that pressing on the top of the screen could cause the screen to tilt backwards, the primary navigation elements were moved to the bottom of the window to minimize this likelihood. The dashboard-mounted notebook PC configuration also meant that traditional left-aligned navigation would also be problematic for right-hand use because the user’s arm would obscure visibility of much of the screen during operation. Once again, navigation controls were relocated to minimize this potential issue. Finally, the fact that officers would at times be using the interface with a gloved hand meant that the interface, originally designed for mouse clicks and typical fingertips, was modified to accommodate the oversized digit (see Figure 4) (Herchenrader, personal communication, 23 July 2013).

The remaining primary goal of the project also proved to be its greatest challenge: the need to minimize the amount of data transferred to the MDT. With most Geocortex® implementations, the viewer is configured to dynamically request all data from a web-connected map server. A major component of data transfer is any tiled map services (typically, this refers to the base maps or related aerial imagery that web-mapping applications provide to orient the user). This normally consists of large

Figure 4. VPD interface.

144 T. Herchenrader and S. Myhill-Jones

data files that contain static map data. To avoid transmitting this large amount of data over the cellular network whenever the viewer was refreshed, the Geocortex® application was modified to load the base map tile package directly to the laptop while still at the office. While conceptually simple, this change required modification to the pre-built viewer via programming and other non-routine configuration. The challenge associated with this undertaking was compounded by the requirement that all testing be completed in the secure VPD environment, rather than at the development laboratory (Herchenrader, personal communication, 23 July 2013).

When the second-generation GeoDash application is implemented in the police vehi- cles, VPD officers will have more ubiquitous access to data and crime analysis capabili- ties. Once a user-specified crime analysis query is completed, the results are presented alongside a standardized set of charts (see Figure 5).

A benefit of this simple, pre-defined interface and analysis capability ensures that users require little to no training in order to submit relevant crime analysis queries, create charts, and view pertinent crime data. It was recognized that the MDT experience needed to be appealing and intuitive in its design to attract and retain users (Herchenrader, personal communication, 23 July 2013). Removing barriers to real-world use and making it as easy as possible for officers to access relevant data and crime analysis serves the overarching goal of helping VPD become more proactive in their policing workflows.

Like WRPS, the VPD is also committed to providing the general public with more crime data and basic analysis tools by making available a public-facing crime data application in early 2014. To avoid any requirement for the public to have a browser

Figure 5. VPD charting options and result clustering.

Police Practice and Research: An International Journal 145

plug-in installed on desktops or a native app installed on tablets and smartphones, and to generally maximize access across traditional PCs, tablets, and smartphones, the VPD decided to implement their public-facing crime map using a still-maturing HTML5 viewer approach.

In the future, with an effort that joins community awareness with intelligence-led policing, the VPD plans to introduce a reporting feature that will allow citizens to report suspicious activity in their neighborhood via the public-facing crime-mapping applica- tion. This differs from the WRPS application which allows citizens to provide further information about occurrences that have already been reported (Herchenrader, personal communication, 13 August 2013). The ability to access community-provided informa- tion on suspicious activities would provide VPD officers with another data source to inform their policing decisions and opportunities to operate preventatively.

Continuing to look towards the future, the VPD is working with Latitude Geographics and the Spatial Pattern Analysis and Research Laboratory at the University of Victoria to develop a specialized property crime-prediction algorithm to further enhance the internal crime-mapping application. This algorithm is being developed for use against property crimes such as car theft and burglary, which tend to be more predict- able than crimes against people (Herchenrader, personal communication, 13 August 2013). In order to test the validity of the prediction algorithm, the VPD plans to imple- ment a pilot program, which will place officers in locations where a crime is predicted to occur to validate if there is a net reduction in crime at those locations (Herchenrader, per- sonal communication, 13 August 2013). Previous studies have indicated that police pres- ence in crime hotspots results in a decrease in crime (Bayley & Garofalo, n.d.; Koper, 1995; Sherman & Weisburd, n.d.). This trial is planned to test the predictive capability of the algorithm as well as to determine whether a more targeted police presence will pro- duce similar crime reduction results. A challenge will likely be allocating sufficient resources towards putting the officers in these locations, as this method remains unvali- dated in Vancouver (Herchenrader, personal communication, 13 August 2013).

Conclusion

Responding to crime will undoubtedly continue to be a core focus for police. However, especially when faced with tighter budgets and increasing demands for public account- ability, police departments are working harder than ever to do more with less. Web-based crime-mapping and analysis capabilities provide considerable improvements over previous paper-based methods of information dissemination, with the benefits of a COTS software application, allowing agencies to quickly and efficiently focus their resources on custom requirements. Using modern technology to implement web-based crime-mapping and analysis capabilities at relatively low cost and risk can empower officers to target crime more proactively while on patrol, offering real potential to mark- edly increase their overall effectiveness. Additionally, providing anonymized crime data to the public satisfies increased expectations for transparency while introducing exciting potential for collecting tips and information from the community to further enhance intelligence-led policing. It is in this context that proactive approaches to policing, such as intelligence-led policing, as well as public-facing crime maps, have gained promi- nence in recent years. With technologies such as GIS able to reach more stakeholders, law enforcement now has the capability to not only better understand spatial patterns related to historic crime, but also consider, in a more informed manner, where they might occur in the future – and allocate the service’s limited resources more efficiently.

146 T. Herchenrader and S. Myhill-Jones

Notes on contributors Tegan Herchenrader is a technical writer and training coordinator at Latitude Geographics. She graduated with BA honours in Arts and Business at the University of Waterloo in 2010 and later received her MA in Globalization from McMaster University in 2012. She started working for Latitude Geographics shortly after her graduation, assisting in the establishment of their Ontario office and development of their online training program. An active outdoorswoman, she can often be found hiking or horseback riding the region with her dog, or immersed in tending to her vege- table garden.

Steven Myhill-Jones (BSc. Geography) is the founder and CEO of Latitude Geographics, an ESRI® Platinum Partner headquartered in Victoria, Canada. He established and spearheaded the growth of Geocortex®, which is a software technology that helps organizations around the world accomplish even more with ESRI® web-mapping technology. Since 1999, Latitude has assisted with developing and implementing hundreds of web-mapping applications for clients across sectors, including public safety.

References Bayley, D., & Garofalo, J. (n.d.). The management of violence by police patrol officers. Criminol-

ogy, 27(1), 1–26. Friedman, W., Gordon, A., & Maltz, M. (2000). Mapping crime in its community setting: Event

geography analysis. New York, NY: Springer-Verlag. Retrieved from http://books.google.ca/ books?hl=en&lr=&id=XRSjd1RxeAYC&oi=fnd&pg=PR9&dq=benefitsofmappingcrime& ots=rmlipqCwzd&sig=TizFPe7OZGrbnnpY2ciQa2XOOSo

Gotway, C. A., & Schabenberger, O. (2009). Statistical methods for spatial data analysis. Taylor & Francis.

Koper, C. (1995). Just enough police presence: Reducing crime and disorderly behaviour by optimizing patrol time in crime hot spots. Justice Quarterly, 12(4), 649–672.

Malleson, N. (2011). Crime analysis – Exploiting geospatial datasets. GIS Professional, 43, 25–26. Manning, P. K. (2001). Technology’s ways: Information technology, crime analysis and the ratio-

nalizing of policing. Criminology and Criminal Justice, 1, 83–103. Ratcliffe, J. (2002). Damned if you don’t, damned if you do: Crime mapping and its implications

in the real world. Policing and Society, 12, 211–225. Retrieved from http://jratcliffe.net/ papers/Ratcliffe(2002) Damned if you don’t damned if you do.pdf

Ratcliffe, J. (2008). Intelligence led policing. Cullompton: Willan. Ratcliffe, J. (n.d.). What is intelligence led policing? Retrieved from http://jratcliffe.net/research/

ilp.htm Rich, T. F. (1995, July). The use of computerized mapping in crime control and prevention

programs. Retrieved from https://www.ncjrs.gov/txtfiles/riamap.txt Sherman, L. W., & Weisburd, D. (n.d.). General deterrent of police patrol in ‘crime hotspots’: A

randomized controlled trial. Justice Quarterly, 12(4). Retrieved from http://www.houseof mouse.ca/sites/default/files/Sherman-General deterrent effects of police patrol in crime-1995. pdf

Police Practice and Research: An International Journal 147

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.