By Joel M. Caplan, Leslie W. Kennedy, Eric L. Piza and Phillip Marotta
There are several ways to make sense of the forces that affect the locations or spatial patterns of crime and, ultimately, create risky places. Evaluating the “spatial influences” of features of the landscape on the occurrence of crime incidents, and assessing the importance of each feature relative to one another is a viable method of assessing such risk (Caplan, 2011). Spatial influence refers to the way in which features of a landscape affect places throughout the landscape. For an example that is much more benign than criminal offending, consider a place where children repeatedly play. When we step back from our focus on the cluster of children, we might realize that located where they play repeatedly exists swings, slides and open fields. These features of the place (i.e. suggestive of a playground) attract children instead of other locations that lack such entertaining qualities. Features of a landscape can influence and enable playful behaviors. In a similar way, spatial factors can influence the seriousness and longevity of illegal behaviors and associated crime problems (Caplan, Kennedy, & Piza, 2012). Risk terrain modeling (RTM) identifies the risks that come from features of a landscape and models how they co-locate to create unique behavior settings for crime (Caplan & Kennedy, 2010).
Risk Terrain Modeling
RTM is an approach to risk assessment whereby separate map layers representing the spatial influence of features of a landscape are created in a geographic information system (GIS). Then risk map layers of statistically validated features are combined to produce a composite “risk terrain” map with values that account for the spatial influences of all features at every place throughout the landscape. Within the context of RTM, modeling refers to the process of attributing qualities of the real world to places throughout a landscape, and combining multiple landscapes together to produce a single composite map where the newly derived value of each place represents the compounded risk of that place. RTM offers a statistically valid way to articulate crime-prone areas at the micro‐level according to the spatial influence of many features of the landscape, such as bars, parks, schools, ATMs, or fast food restaurants. Risk values in a risk terrain model do not create absolute scenarios where crimes will ensue. They simply point to locations where, if the conditions are right, the risk of illegal behavior will be high.
Risk terrain modeling was developed at the Rutgers University School of Criminal Justice where resources for performing RTM are freely available online (see www.rutgerscps.org). Risk terrain modeling is not difficult. Many police agencies are using it on a regular basis. But, to make it more accessible for people with limited GIS and statistical abilities, Rutgers developed the Risk Terrain Modeling Diagnostics (RTMDx) Utility, a software app that automates RTM (also available for free online). The RTMDx Utility diagnoses spatial attractors of crime and communicates actionable information to identify where new crime incidents will emerge or cluster, and to strategically and tactically allocate resources (Caplan, Kennedy, & Piza, 2013). There are no ‘black boxes’ to the analytical and statistical methods in the RTMDx Utility, and all details are included in the user manual for replication and manual production. The Utility is simply intended to help automate the RTM process, not hide it.
Strategic Example of RTM
Current strategic applications of risk terrain modeling are exemplified by ongoing projects in seven cities across the United States that are funded by the National Institute of Justice (NIJ) and led by our team at Rutgers University. Partner police departments include New York, NY; Newark, NJ; Chicago, IL; Kansas City, MO; Arlington, TX; Colorado Springs, CO; and Glendale, AZ. A key objective of the projects is to inform police-led interventions that address a designated priority crime type at target areas, respectively, for each city. Using the RTMDx Utility, RTM diagnoses the underlying spatial factors of crime at existing high-crime places. Then activities are designed to suppress crime in the short-term and mitigate spatial risk factors at these areas to make them less attractive to criminals in the long-term.
Outcome evaluations of the impacts of these risk-based interventions are still ongoing, and it is too early to discuss results at this time. But, a preliminary process evaluation suggests that intel produced from risk terrain modeling was pragmatic and meaningful for police officers when developing and implementing risk-based intervention strategies at target areas. RTM products also directly informed steps to mitigate spatial risks and advance overall short- and long-term objectives. The objectives were to constructively change the frequencies and spatial distributions of crime events at micro-places. Notably, the interventions were designed in a manner that did not place an undue burden on police department resources or finances. The intervention strategies are considered to be reasonably sustainable and repeatable under “normal” (i.e., non grant-funded) conditions. This is especially important if they are proven to yield statistically significant outcomes.
Tactical Example of RTM
A tactical example of risk terrain modeling was recently inspired by the federal government’s challenge to reduce injuries and fatalities among law enforcement officers (LEOs). In the past several decades, research has examined situational, offender, and individual characteristics of the risks of non-accidental injury and death to LEOs in the line-of-duty. But the influence of specific features of the physical environment has remained largely absent from empirical study. With data obtained from the Chicago Police Department, we used RTM to identify features of the physical landscape that constitute significantly higher risk of felonious assault/battery to police officers handling calls-for-service.
In 2012, there were 919 batteries and 72 assaults with a firearm against Chicago police officers. In order of relative risk values, the spatial risk factors of these incidents are: foreclosures, problem buildings, bars, schools, gang territories, banks, apartment complexes, liquor stores, 311 service requests for street lights all out, grocery stores, and retail shops. All places may pose risk of assault/battery to officers when dealing with a variety of types of calls-for-service, but because of the spatial influence of certain features of the landscape, some places are riskier than others.
Places where more than one of these features co-locates pose even higher risks. Relative risk values for each micro place (e.g., a street block) in the risk terrain map shown in Figure 1 ranged from 1 for the lowest risk place to 582.5 for the highest risk place. A place with a value of 582.5 has an expected rate of assault/battery to police that is 582.5 times higher than a place with a value of 1. The mean risk value is 15.33, with a standard deviation of 23.60. This micro level map shows the highest risk places symbolized in black (i.e., greater than 2 standard deviations from the mean risk value). The likelihood of experiencing assault/battery on police officers at the highest risk places are 62.53% greater than the risk presented to police officers managing calls-for-service at other locations. Albeit, there are many other factors that could be taken into account to assess personal risks to officers responding to calls-for-service. For instance, additional research is needed to assess the temporal dynamics of assault/battery incidents, as well as the situational factors (i.e., uniform or plain-clothes officer, multi-person/car first responders, etc.). Within the scope of RTM, though, it can be said with statistical confidence that such events occur at places with particular features of the landscape.
Predicting crime is a tall order. We are not at the point where we can predict specific events by particular offenders at certain moments in time. But, with RTM we can identify the most vulnerable areas in a jurisdiction which allows us to predict, with a certain level of confidence, the most likely places where crimes will emerge—even if they haven’t occurred there already. RTM is being used to help explain why spatial patterns of crime exist in a jurisdiction, and what can be done to mitigate risks, not just chase the “hotspots.” Certain features of the built environment can increase the risk of crime, and places with high relative risk values as defined by a risk terrain model are behavior settings that present exceptionally strong likelihoods of criminal events.
Utilizing environmental factors for crime analysis has many benefits. One benefit is an emphasis on intervention activities that focus on places, not just people located at certain places – which could jeopardize public perceptions and community relations. Another is that RTM is a sustainable technique because it does not need past crime data to make valid forecasts. Our researcher-practitioner collaborations, using RTM, have led to new approaches to police productivity that go beyond a heavy reliance on traditional law enforcement actions such as stops, arrests or citations. Police departments around the country are using RTM to be problem-oriented and proactive, to prevent new crimes without concern that a high success rate (and no new data) will hamper their ability to make new forecasts.
RTM contributes to the law enforcement mission by providing evidence-based spatial intelligence that police can employ to mitigate the risk of crime and violence at micro places throughout urban, suburban or rural jurisdictions. Our research suggests that there are empirically important spatial risk factors whose presence or absence structures the potential for crime to emerge and cluster. RTM provides information for police agencies and other stakeholders to incorporate into actions that yield meaningful and measurable results.
Caplan, J. M. (2011). Mapping the spatial influence of crime correlates: A comparison of operationalization schemes and implications for crime analysis and criminal justice practice. Cityscape, 13(3), 57-83. Retrieved March 1, 2014 from http://www.huduser.org/portal/periodicals/cityscpe/vol13num3/Cityscape_Nov2011_mapping_spatial.pdf
Caplan, J. M. & Kennedy, L. W. (2010). Risk Terrain Modeling Manual: Theoretical Framework and Technical Steps of Spatial Risk Assessment. Retrieved March 1, 2014 from www.rutgerscps.org/rtm
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. Retrieved March 1, 2014 from http://rutgerscps.org/software
Kennedy, L. W., Caplan, J. M. & Piza, E. L. (2012). A Primer on the Spatial Dynamics of Crime Emergence and Persistence. Retrieved March 1, 2014 from http://www.rutgerscps.org/PrimerOnCrimeBook.htm
Joel M. Caplan, Ph.D., is an Associate Professor at Rutgers University School of Criminal Justice and Deputy Director of the Rutgers Center on Public Security, where he co-developed risk terrain modeling methods for crime analysis. His research focuses on risk assessment, spatial analysis, and computational criminology, which takes the strengths of several disciplines and builds new methods and techniques for the analysis of crime and crime patterns. Joel has professional experience as a police officer, 911 dispatcher, and emergency medical technician.
Leslie W. Kennedy, Ph.D., is a University Professor at Rutgers University School of Criminal Justice (SJC) and Director of the Rutgers Center on Public Security. In his most recent research, he has focused on developing, with Joel Caplan, Risk Terrain Modeling, a GIS based analytical program, for use by law enforcement in forecasting and preventing crime. He has published in major criminology journals, including Criminology, Justice Quarterly, and the Journal of Quantitative Criminology.
Eric L. Piza, Ph.D., is an Assistant Professor at John Jay College of Criminal Justice, Department of Law and Police Science. He has over 10 years of crime analysis and program evaluation experience, with previous professional positions at the Newark Police Department, Rutgers School of Criminal Justice, Rutgers Center on Public Security, and the Police Institute. Dr. Piza is currently involved in a number of applied research projects in partnership with public safety agencies across the United States.
Phillip Marotta is a doctoral student In Criminal Justice at Rutgers University. His research interests focus on encounters between law enforcement and persons with mental illness, Crisis Intervention Team Programs, emergency psychiatry, violence prevention and quantitative methods. Mr. Marotta has masters degrees in Clinical Social Work and in Public Health, both from Columbia University. Before coming to Rutgers University, Mr. Marotta completed a one-year clinical fellowship through the Department of Psychiatry at Yale University.