Below you will find tools and ideas on using the Evidence-Based Policing Matrix.  Please use the quick links below to find out more information.  Looking for information on the Matrix axes and coding? Click here for the Matrix Key.

Articles on the use of the Matrix
The Matrix Demonstration Project
Video Training
Other Ideas – for Police Agencies
Other Ideas – for Researchers and Funding Agencies


Articles on the use of the Matrix

  • Lum, C. (2009). Translating Police Research into Practice. Ideas in American Policing Series. Washington, DC: Police Foundation.
  • Lum, C., Koper, C., and Telep, C. (2011). The Evidence-Based Policing Matrix. Journal of Experimental Criminology 7(1): 3-26.
  • Lum, C., Telep, C., Koper, C., and Grieco, J. (2012). Receptivity to Research in Policing. Justice Research and Policy 14(1).
  • Lum, Cynthia and Christopher Koper. (forthcoming). Evidence-Based Policing. In The Encyclopedia of Criminology and Criminal Justice (Gerben Bruinsma and David Weisburd, Editors). Springer-Verlag.

The Matrix Demonstration Project

The Matrix Demonstration Project provides examples of institutionalizing principles from the Matrix and also from police research more generally into practice. Agencies can download ideas, forms, guidelines, and videos from the MDP to develop their own efforts.


Video Training on Evidence-Based Policing and the Matrix

Other Ideas – For Police Agencies

1. Officers in Patrol or Street-Level Units:

The Matrix generally indicates that patrol can improve its crime control effectiveness by balancing their reactive, rapid-response 911 approach, with proactivity at crime “hot spots” or problem places in between calls for service. Roughly between 40-60% of an officer’s time, even in high crime cities, is not committed to answering calls for service. The Matrix (and many others, including the POP Center, the COPS Office, or provides examples and ideas of directed patrol, problem-solving, and community-building ideas for patrol officers to consider. Further, the authors recommend applying the Koper Curve Principle in crime hot spots; that officers and units randomly hit hot spots and stay for limited amounts of time to maximize their deterrent effect.

Some general patrol deterrent strategies at places in the past have included crackdowns, proactive traffic stops, and arrest policies for disorder and quality of life crimes. More recently, technologies have helped facilitate general deterrent patrol in hot spots, including license plate readers and CCTVs. And, a number of problem-solving activities which extend partnerships to non-police and other community entities have also been developed. Officers should be cognizant, and sensitive (and consistently trained to be so) to concerns that some crackdown approaches can also result in racial profiling, excessive use of force, reductions of police legitimacy, and overcrowded jails. Educating and alerting communities about these strategies and working to professionally and constitutionally deliver proactive policing at places is an important part of implementing police tactics.

The Matrix also recommends problem solving and multi-agency approaches at smaller places. This can include civil remedies (nuisance abatement) such as those shown in Mazerolle’s studies (e.g. Mazerolle, Price, & Roehl, 2000) problem-oriented policing approaches as seen in the micro place slab of the Matrix, “Project Safe Neighborhood” approaches and other more specific community-level tactics. Another promising approach may be a “pulling levers” approach with regard to gang violence at places. Officers are encouraged to examine specific abstracts/summaries of the studies within these realms that have shown promise (the black dots) for further ideas.

2. Strategizing about Tactical Operations and Patrol Deployment:

These suggestions are similar to the uniformed officer discussion above, and should be read in addition to #1. Clusters of effective studies (indicated by the black dotes) indicate realms of effectiveness within the Matrix. Rather than try to apply a single study to an agency’s more specific concerns, these realms provide clues as to which three intersecting dimensions have clusters of effective studies within them.

For example, where the dimensions “micro place”, “focused” and “proactive” or “highly proactive” intersect, there are a number of studies which indicate interventions like these can be fruitful in reducing crime. Agencies developing crime prevention measures need not replicate a study’s specific intervention. Rather, the Matrix shows that if agencies develop patrol interventions that are aimed at micro-places, focused/tailored, and proactive, then they may be able to improve their ability to reduce crime. Agencies that undertake reactive and individual-based approaches may not be as productive. Thus, a strategy involving nuisance abatement of a house which is used to distribute drugs is more likely to be effective in reducing drug crimes than building a case against a street dealer using buy and bust or other traditional techniques.

The realms of effectiveness in the Matrix are a research translation tool for strategic development; rather than search for a study that is specifically relevant to a jurisdiction, the clustering of effective studies helps create generalizations (intersection dimensions) which can provide enough flexibility for jurisdiction-specific approaches, but at the same time are informed by evidence. Devising an overall strategy for patrol, specialized or even investigative units that is “housed” in a realm of effectiveness will prove more fruitful than a “random” patrol strategy at beats that is built from goals that are purely procedures-based.

3. Assessment and accountability at the command and agency level:

The Matrix can be interacted with at command meetings such as COMPSTAT or other managerial practices to determine if the most effective approaches are being used for a particular crime problem and also to foster evidence-based leadership in general deployment. For example, during a COMPSTAT meeting, the focus could be shifted from reciting monthly statistics and vague assertions of tactics to real-time mapping of tactical suggestions and existing strategies directly into the Matrix, discussing what alterations in tactics and strategies can push approaches towards more effective realms.

The Matrix could be also more broadly used to self-assess the agency’s overall tactical portfolio. Mapping an agency’s existing patrol interventions and tactics into the Matrix can provide a comparison between the tendencies of an agency’s patrol deployment and what is known from research.

4. Assessment and accountability at the first line-supervision and district levels:

Such assessments of tactical and strategic portfolios could also be carried out at the district, sector, squad, or individual level. In the U.S., first line supervisors (sergeants) have had more passive supervisory duties, such as approving reports, applying general orders to infractions, violations, or special concerns, handling administrative duties regarding personnel, or responding with officers to serious calls for service. Empowering first line supervisors with responsibility in developing their own tactics and strategies for their squads can increase the motivation and effectiveness of this rank, moving towards a more active model of first line supervision. The Matrix provides a quick visual guide for first line and district supervisors to determine whether their choices most likely will lead to positive results. It can also influence the use of crime analysis at the patrol and first-line supervisory level, given that most effective tactics require basic crime analysis. This is an essential step forward toward not only integrating crime analysis at the street-level, but also increasing the proactive tendencies of an agency.

5. Early training and socialization of officers:

Along these same lines, officers should be trained during the academy and in-service courses in the fundamental concepts of how to increase their effectiveness and legitimacy by exposing them to tactics (or more generally, realms of effectiveness) backed by scientific evidence, not by anecdotes, stories, or other ad hoc experiences. A basic skill of the entering officer should be the ability to identify which tactics have been shown to be effective in reducing crime and which have not, just as he or she also learns the procedures by which to make an arrest. Further, tactics in the most effective realms of the Matrix emphasize other approaches to policing than just reacting to 911 calls or cases on an individual basis. Transitioning police culture and mentality away from one that is reactive to one that is proactive is a fundamental need for early training as well as in-service education. The Matrix Demonstration Project web page has examples of how this might occur.

6. Promotions and advancement:

Candidates, when tested on crime prevention scenarios (which is often common at the first and second-level supervisor ranks), could be assessed on their ability to develop solutions that fall within effective realms. Or, the tactical resumes/portfolios of those in line for promotion could be mapped into the Matrix to discern whether contenders generally use approaches that are more evidence-based or if they tend to rely on methods that are more traditional.

7. Crime analysis units:

Crime analysts are the “applied criminologists” of agencies. They have the opportunity not only to assess the crime control and preventative effects of police tactics, but also to use the most believable scientific methods in their evaluations. The Matrix, especially the tab showing the Matrix “divided by rigor” provides ideas on evaluating new strategies using experimental methods, and ideas on what other analysts and researchers are doing to assess the effectiveness of tactics.

Further, the authors recommend agencies increase their analysis units if they take evidence-based approaches. The most effective realms of the Matrix are those that fall in place-based, proactive areas, requiring constant computerized crime mapping support for operational units, as well as crime analysis to discover crime trends for proactive deployment or predictive policing.


Other Ideas – For Research and Funding Agencies

1. Supplying methodologically rigorous research to meet the demand for research:

The Matrix not only shows researchers and funding agencies areas of police intervention that have not been researched but also areas that have not been researched well. To make this point, the Matrix is split below into two groups of studies. Figure A represents 71 studies (or 61% of the entire Matrix) that used moderately rigorous designs, while Figure B shows the 46 studies (39%) that used stronger methods. This separation shows that there are much fewer evaluations available of high quality. Thus, while we know some areas of the Matrix are “promising”, more is needed on being more convinced of this promise.

Comparisons of Studies in the Matrix of Moderate and Strong Methods:




A few other findings stand out in this division – higher quality studies that show positive effects are most consistently found in the proactive, micro place-based region. Figure B also tells us, with high certainty, that individual strategies are much less promising, and in some cases, harmful. Finally, notice that neighborhood and group-based studies almost completely disappear when only looking at the highest quality studies. If police are interested in using such strategies, then better information must be generated at these units of analysis. For example, the policing of gangs is a high priority issue for police, yet very little strong evaluation research exists in the “groups” slab of the Matrix to meet this demand for evaluation.

2. A common language for the research-practitioner relationship:

The Matrix also can be used as a “common ground” for conversations between researchers, police practitioners, and funding agencies when collaborating to evaluate, study, and ultimately reduce crime. In many ways, the Matrix builds on officer experience by connecting to officers with familiar vernacular. For example, a police agency may be interested in testing certain types of interventions such as crackdowns on gangs or illegal gun carrying. The researcher, however, may be interested in improving the quantity of high quality evaluations in the proactive place-based regions of the Matrix or conducting more rigorous experiments of neighborhood-level policing. In this scenario, the Matrix could be used to elicit discussion and negotiation between the researcher and the police agency in a way that keeps the agency grounded in evidence-based regions but that does not divorce the police researcher from the real needs of the police agency. Solutions might thus include a quasi-experimental study testing pulling levers approaches in multiple gang territories, or perhaps a randomized repeated measures study of crackdowns on gun carrying in high-risk patrol beats.

3. Evidence-based funding and “Smart Policing” efforts:

Funding agencies, such as NIJ, the BJA, or the COPS Office, can use tools like the Matrix to fund research and interventions in strategic ways that facilitate evidence-based policing, thus placing these agencies in more of a leadership role regarding improving the use of evidence in law enforcement. As shown in the figure below, such agencies might give priority, for example, to “low risk” funding that would support more research in understanding the nature of interventions that work, or in testing replications in diverse areas (for example – do place-based approaches work – and how – in rural or suburban areas as they do in urban areas?)

“Medium risk” areas of the Matrix point to intersection dimensions in which the research may be less rigorous, or findings might be more ambiguous. More rigorous research in these areas may be needed, or research that examines what aspects of these interventions lend to crime prevention.

“High risk” programs and research would fall within domains of the Matrix that have shown little promise or even backfire effects. Here, researchers and funding agencies might show constraint in funding within these areas. Or, if funding is necessary, focus on better understanding when such approaches might be used effectively.

Finally, there are areas of the Matrix that reflect high demand for research, but where the supply of research is low. Evaluations of gang and co-offending interventions, or jurisdiction-level policies may be good targets for innovative or exploratory research and research funding.