Police Methods | Section 5.4

Police Methods

Adam J. McKee, Ph.D.

Section 5.4: Assessment and Response

The excerpt below was written by Ronald V. Clarke & John E. Eck.


Crime Analysis For Problem Solvers In 60 Small Steps

7. Be guided by SARA — but not led astray!

Within problem-oriented policing, the police are required to: (1) carefully define specific problems (see Step 14 for the definition of “problem”); (2) conduct in-depth analyses to understand their causes; (3) undertake broad searches for solutions to remove these causes and bring about lasting reductions in problems; and (4) evaluate how successful these activities have been. This is a form of action research, a well-established social science method in which researchers work alongside practitioners, helping to formulate and refine interventions until success is achieved. This can be contrasted with the usual role of researchers, in which they work apart from the practitioners, collect background information about problems, and conduct independent evaluations. In action research, however, the researcher is an integral member of the problem-solving team. This is the role of the crime analyst. Your analyses must inform and guide action at every stage.

You will find that SARA will help you and your team keep on track. This is the acronym formulated by John Eck and Bill Spelman to refer to the four problem-solving stages of Scanning, Analysis, Response and Assessment. This process is very similar to many other analytical processes, including the standard crime analysis process of collection, collation, analysis, dissemination, and feedback. By dividing the overall project into separate stages, SARA helps to ensure that the necessary steps are undertaken in proper sequence – for example, that solutions are not adopted before an analysis of the problem has been undertaken. This is a useful check on the natural tendency to jump straight to a final response, while skimping on definition of the problem and analysis and forgetting to assess their impact on the problem.

Problem-solving projects can be complex. In action research, the team is expected to persist until success is achieved, refining and improving an intervention in the light of what is learned from earlier experiences. The process is not necessarily completed once the assessment has been made. If the problem persists, or has changed its form, the team may have to start over. This is represented in the figure where the outer arrows describe the feedback between assessment and scanning.

However, the four problem-solving stages do not always follow one another in a strictly linear fashion. In fact, projects rarely follow a linear path from the initial scanning and analysis stages through the stages of response and assessment. Rather, the process often has loops, so that an unfolding analysis can result in refocusing of the project, and questions about possible responses can lead to the need for fresh analyses. The longer and more complicated the project, the more loops of this kind are likely to occur. The set of smaller inner arrows in the figure illustrate this dynamic process. For example, one might jump from scanning to the implementation of a short-term emergency response to stabilize the problem while further analysis is undertaken. An assessment of the short-term response could add to the analysis and contribute to the formulation of a new response, which is then assessed. This might lead back to scanning as new information forces a revision of the problem definition or the discovery of new problems. The important point is that analysis and evaluation are meaningfully incorporated into the sequence of events and one does not simply jump from scanning to response and declare victory.

One of us (Clarke) recently worked with Herman Goldstein on a project to reduce thefts of appliances from houses under construction in Charlotte, North Carolina. The housing developments were often in fairly isolated rural areas and were impossible to patrol effectively. They were difficult to secure because builders wanted to encourage prospective buyers to tour the sites in the evenings and on weekends. Because few offenders were ever caught, we knew little about them or how they disposed of the appliances. We considered a wide range of possible solutions including storing appliances in secure containers on site and the use of portable alarms and closed-circuit television cameras. Then we hit on a solution being used by some small builders – to delay installation of the appliances until the day that the buyer took possession. 

Many builders were hostile at first to the idea. Sales staff believed that having the appliances installed made a home more saleable, and that the absence of appliances, if attributed to theft, might alarm purchasers about the area they were moving into. Site supervisors felt that delivering and installing appliances as houses were occupied would be more difficult than batch delivery and installation. Some erroneously believed that building inspectors would not certify the houses as suitable for occupancy unless appliances were in place. Others wrongly believed that this was a mortgage requirement. Finally, individual installation would mean that builders could no longer arrange for building inspectors to visit a site and issue certificates of occupancy wholesale. 

Because the solution had so many advantages, we decided to return to the analysis stage to find answers to the builders’ objections. Ultimately, this information was useful in persuading builders to adopt the solution and thus reduce the number of appliance thefts. 

14. Use the CHEERS test when defining problems

A problem is a recurring set of related harmful events in a community that members of the public expect the police to address. This definition draws attention to the six required elements of a problem: Community; Harm; Expectation; Events; Recurring; and Similarity. These elements are captured by the acronym CHEERS:


Community. Members of the public must experience the harmful events. They include individuals, businesses, government agencies, and other groups. Only some – not all or most – community members need experience the problem.

Harmful. People or institutions must suffer harm. The harm can involve property loss or damage, injury or death, serious mental anguish, or undermining the capacity of the police (as in repeat fraudulent calls for service). Illegality is not a defining characteristic of problems. Some problems involve legal behavior that the police must address. Noise complaints arising from the impact of legitimate commercial activity on neighboring residents is a common example. Some problems are first reported as involving illegal behavior, but on closer inspection do not involve illegalities. If such reports meet all the CHEERS criteria, they are problems.

Expectation. Some members of the community must expect the police to address the causes of the harm (their numbers do not have to be large). Expectation should never be presumed, but must be evident through processes such as citizen calls, community meetings, press reports, or other means. This element does not require the police to accept at face value the public’s definition of the problem, their idea of its causes, or what should be done about it. The public may be mistaken as to its cause and characteristics. It is the role of analysis to uncover the causes.

Events. You must be able to describe the type of event that makes up the problem. Problems are made up of discrete events. Examples of events include a break-in at a home, one person striking another, two people exchanging money and sex, or a burst of noise. Most events are brief, though some may involve a great deal of time – some frauds, for example.

Recurring. These events must recur. Recurrence may be symptomatic of acute troubles or a chronic problem. Acute troubles suddenly appear, as in the case of a neighborhood with few vehicle break-ins suddenly having many such break-ins. Some acute troubles dissipate quickly, even if nothing is done. Others can become chronic problems if not addressed. For this reason, acute troubles should be investigated to determine if they signal something more entrenched. Chronic problems persist for a long time, as in the case of a prostitution stroll that has been located along one street for many years. Unless something is done, the events from chronic problems will continue to occur.

Similarity. The recurring events must have something in common. They may be committed by the same person, happen to the same type of victim, occur in the same types of locations, take place in similar circumstances, involve the same type of weapon, or have one or more other factors in common. Without common features, you have an arbitrary collection of events, not a problem. Common crime classifications – such as used by the Uniform Crime Reports – are not helpful. Vehicle theft, for example, includes joyriding, thefts for chop shops, thefts for export to other countries, thefts for use in other crimes, and a host of other dissimilar events. So a cluster of vehicle thefts may not be a single problem. More information is needed. With common features, we have a pattern of events that could indicate a problem – for example, thefts of minivans in suburban neighborhoods to be used as gypsy cabs in the inner city.

Problems need to be examined with great specificity (see Steps 6 and 15) because small details can make a difference between a set of circumstances that gives rise to harmful events, and a set of circumstances producing harmless events. CHEERS suggests six basic questions you need to answer at the scanning stage:

  • Who in the community is affected by the problem?
  • What are the harms created by the problem?
  • What are the expectations for the police response?
  • What types of events contribute to the problem?
  • How often do these events recur?
  • How are the events similar?

Not everything the police are asked to address is a problem. CHEERS can help identify demands that are not problems. We are using the term “problem” in the technical, POP sense, not as we would in everyday speech. So things that are not problems may be troublesome and may require police attention. These are as follows:

Single events. A single event, regardless of how serious, is not a problem unless there is a reasonable prospect that another similar event will occur if nothing is done. A single event may deserve investigation or some other police action, but problem solving cannot be applied to isolated events because nothing can be prevented.

Neighborhoods. Small areas, such as city centers or particular residential apartment complexes, sometimes get reputations as problems, but these neighborhoods are seldom problems. Rather they are usually areas containing several problems. The individual problems might be related, but not always. Tackling an entire area as a single problem increases the complexity of the effort and reduces the chances you will find effective responses. Instead, you should identify specific problems within a neighborhood and tackle them individually. If the problems are linked (e.g., the street network contributes to several problems) then tackling the link might be helpful. Do not assume problems are linked just because they are near each other. In some cases, of course, there may be common solutions to distinct problems (see Step 6).

Status conditions. Truant schoolchildren, bored teenagers, vagrant adults, and convicted criminals are not problems because of their status of not being in school, having nothing to do, not being employed, or having been found guilty of an offense. A community might expect the police to do something about them, but status conditions lack the characteristics of harm and events. Some of these people may play a role in problems, as targets, offenders, or in some other capacity, but that does not make them a problem. Defining a problem by status conditions is evidence of lack of precision and a need to examine the issue in greater depth. Status conditions may point to pieces of a larger problem.

15. Know what kind of problem you have

Because local police have to deal with a wide range of problems that meet the CHEERS definition (Step 14) we have developed a classification for these problems. This classification scheme can help you precisely define the problem. It helps separate superficially similar problems that are really distinct. It also allows you to compare your problem to similar problems that have already been addressed, and it helps identify important features for examination. For example, an extensive set of guides to addressing common problems are available from the Office of Community Oriented Policing Services and the Center for Problem-Oriented Policing websites (Step 19). Knowing the type of problem you are investigating can help you identify guides that might be helpful, even if they do not directly address your problem. The classification scheme is based on two criteria: the environments within which problems arise, and the behaviors of the participants. (The scheme is different from the wolf/duck/den classification in Step 8, which is a classification of persistent problems.)

Environments regulate the targets available, the activities people can engage in and who controls the location. Specifying an environment allows comparisons of environments with and without the problem. Environments have owners who can be important for solving the problem (see Step 44). There are 11 distinct environments for most common police problems:

  • Residential – Locations where people dwell. Houses, apartments, and hotel rooms are examples. Though most are in fixed locations, a few are mobile, such as recreational vehicles. 
  • Recreational – Places where people go to have a good time. Bars, nightclubs, restaurants, movie theaters, playgrounds, marinas, and parks are examples. 
  • Offices – Locations of white-collar work where there is little face-to-face interaction between the workers and the general public. Government and business facilities are often of this type. Access to these locations is often restricted. 
  • Retail – Places for walk-in or drive-up customer traffic involving monetary transactions. Stores and banks are examples. 
  • Industrial – Locations for processing of goods. Cash transactions are not important activities in these environments and the public is seldom invited. Factories, warehouses, package-sorting facilities are examples.
  • Agricultural – Locations for growing crops and raising animals. 
  • Education – Places of learning or study, including day care centers, schools, universities, libraries, and place of worship. 
  • Human services – Places where people go when something is wrong. Courts, jails, prisons, police stations, hospitals and drug treatment centers are examples. 
  • Public ways – Routes connecting all other environments. Roads and highways, foot-paths and bike trails, and drives and parking facilities are examples. 
  • Transport – Locations for the mass movement of people. These include buses, bus stations and bus stops, airplanes and airports, trains and train stations, ferries and ferry terminals, and ocean liners and piers. 
  • Open/transitional – Areas without consistent or regular designated uses. These differ from parks in that they have not been designated for recreation, though people may use them for this. Transitional areas include abandoned properties and construction sites.

Behavior is the second dimension for classifying a problem. Specifying behaviors helps pinpoint important aspects of harm, intent, and offender-target relationships. There are six types of behavior:

  • Predatory – The offender is clearly distinct from the victim and the victim objects to the offender’s actions. Most common crimes are of this type. Examples include robbery, child abuse, burglary, bullying, and theft.
  • Consensual – The parties involved knowingly and willingly interact. This typically involves some form of transaction. Examples include drug sales, prostitution, and stolen goods sales. Note, however, that assaults on prostitutes are predatory behaviors.
  • Conflicts – Violent interactions involving roughly coequal people who have some pre-existing relationship. Some forms of domestic violence among adults involve this type of behavior, though domestic violence against children and the elderly is classified as predatory because the parties involved are not coequal.
  • Incivilities – Offenders are distinguishable from victims but the victims are spread over a number of individuals and the harms are not serious. Many concerns that are annoying, unsightly, noisy, or disturbing, but do not involve serious property damage or injury fall into this category. Loud parties are an example. Whether vandalism fits in this category depends on the details. Some forms of vandalism are predatory. Some incivilities are troublesome regardless of the environment, while others are only troublesome in specific environments.
  • Endangerment – The offender and the victim are the same person or the offender had no intent to harm the victim. Suicide attempts, drug overdoses, and vehicle crashes are examples.
  • Misuse of police – A category reserved for unwarranted demands on the police service. False reporting of crimes and repeated calling about issues citizens can handle themselves are examples. This is a category of last resort – for use when the sole harm stemming from the behavior is the expenditure of police resources and when none of the other categories fit.


Source:  COPS Office.  Crime Analysis For Problem Solvers In 60 Small Steps.



Modification History

File Created:  08/10/2019

Last Modified:  08/10/2019

[ Back | Content | Next]

This work is licensed under an Open Educational Resource-Quality Master Source (OER-QMS) License.

Open Education Resource--Quality Master Source License


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.