Problem-Oriented Policing

Successful Case Studies

Michael Scott, Ronald Clarke, Michael S. Scott, Ronald V. Clarke

• Publisher: Routledge
• Year: 2020
• ISBN:
  9780429854477
• Edition: 1
• Topic: Droit
• Subtopic: Théorie et pratique du droit

Problem-Oriented Policing: Successful Case Studies is the first systematic and rigorous collection of effective problem-oriented policing projects. It includes more than twenty case studies from among the thousands of projects submitted for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. The volume describes in detail the case studies and explains the wider significance of each for effective, efficient, and equitable policing.

This book explores a wide range of problems that fall under five general categories: gang violence; violence against women; vulnerable people; disorderly places; and theft, robbery, and burglary. The case studies tell stories of how police, in collaboration with others, successfully tackled real-world policing problems fairly and effectively. The authors have also drawn out of the case studies the cross-cutting themes and issues they illustrate. The authors prove that the concept can work, bring to life the context in which police and communities addressed these vexing problems, and, ideally, will inspire future problem-oriented police work that builds on these reported successes.

Written in a clear and direct style, this book will appeal to students and scholars of policing, criminology, and social studies; police practitioners and crime analysts; and all those who are interested in learning more about the reality of police problem-solving.

1
INTRODUCTION

Michael S. Scott and Ronald V. Clarke

Problem-oriented policing is not a new idea; indeed, it has been around for more than forty years with the publication of Herman Goldstein’s (1979) classic article outlining the approach. In that article, Goldstein advised police departments to abandon their usual mode of allocating resources for the narrow objectives of patrolling territory, responding quickly to service requests, and detecting criminal activity and instead to focus more directly on addressing the crime and disorder problems that constitute their business. He urged that policing problems were best addressed by coordinating highly specific societal and environmental changes that would, ideally, prevent the problems or, at a minimum, reduce the harm they were causing the public.

Since then, the benefits and practicality of problem-oriented policing have been analyzed in many different publications (see, e.g., Goldstein, 1990; Scott, 2000; Clarke, 2002; Bullock & Tilley, 2003; Bullock, Erol & Tilley, 2006; Braga, 2010; Scott et al., 2017). In this book, we assume that readers are familiar with the concept and we do not explain it further (however, Box 1.1 contains Goldstein’s own succinct description as a brief reminder). Instead, we write with the different objective of gathering in one place a collection of successful case studies of problem-oriented policing.

BOX 1.1 GOLDSTEIN’S (2001) SUMMARY OF PROBLEM-ORIENTED POLICING

Problem-oriented policing (POP) is an approach to policing in which (1) discrete pieces of police business (each consisting of a cluster of similar incidents, whether crimes or acts of disorder, that the police are expected to handle) are subject to (2) microscopic examination (drawing on the especially honed skills of crime analysts and the accumulated experience of operating field personnel) in hope that what is freshly learned about each problem will lead to discovering a (3) new and more effective strategy for dealing with it. POP places a high value on new responses that are (4) preventive in nature, that are (5) not dependent on the use of the criminal justice system, and that (6) engage other public agencies, the community and the private sector when their involvement has the potential for significantly contributing to the reduction of the problem. POP carries a commitment to (7) implementing the new strategy, (8) rigorously evaluating its effectiveness, and, subsequently, (9) reporting the results in ways that will benefit other police agencies and that will ultimately contribute to (10) building a body of knowledge that supports the further professionalization of the police.

We believe this will be helpful to police, either those who are already engaged in problem-oriented policing but who might learn something from recognized successes or those who are thinking about undertaking it but need proof that it can work. Neither group of officers is likely to have time to search through the hundreds of case studies of problem-oriented policing on the website of the Center for Problem-Oriented Policing (www.popcenter.org), and even if they did, they might not feel confident in judging the success of particular projects. We trust this book will do that work for them.

We also hope the book will be useful to those in the academic world who are teaching or learning about policing. They might know that problem-oriented policing was found to be effective in the systematic review of police research undertaken by the National Research Council (2004) as well as in those undertaken by the Campbell Collaboration (Weisburd et al., 2010) and others (Sherman et al., 1997; Welsh & Farrington, 2002; Telep & Weisburd, 2012). They might also be familiar with the vetting processes of various problem-oriented policing award programs,¹ and they might have heard that Herman Goldstein was awarded the Stockholm Prize in Criminology in 2018 for his work in developing problem-oriented policing. However, we believe this book would also be useful to them as a “go-to” source of detailed accounts of successful case studies that would fill out the dry and abstract conclusions of the systematic reviews mentioned earlier.

In fact, it was Goldstein’s Stockholm Prize that was the immediate stimulus for this volume. The award jury asked one of us, Scott, the director of the Center for Problem-Oriented Policing and a onetime undergraduate student of Goldstein, with a subsequent career in policing, to organize a series of panels at the Stockholm Symposium where the prize is awarded. These panels would be intended to convey to those attending the symposium—many of whom are criminologists who would not necessarily be familiar with work on policing—what problem-oriented policing is and how it is put into practice. Scott, in turn, invited the other of us, Clarke, associate director of the Center for Problem-Oriented Policing, to join him in organizing the panels. Clarke is the leading exponent of situational crime prevention, an approach similar to problem-oriented policing but that has its origins in criminology, not in police administration as is the case for problem-oriented policing. In fact, the links between problem-oriented policing and situational crime prevention have grown stronger in recent years, with much intermingling of scholars (Scott & Goldstein, 2012).

We thought the best way for neophytes to learn about problem-oriented policing would be through presentations of a variety of specific projects undertaken by scholars and practitioners, sometimes working together, and this comprised the panels we organized. Subsequently, we invited symposium presenters to write up their presentations as chapters for this volume to share these examples of effective problem-oriented policing with an audience far wider than was able to attend the Stockholm Symposium.

We do not mean to imply that the case studies in this volume are the very best ever done. We think only that this collection reflects a broad sample of successful problem-oriented policing projects across a wide variety of problem types and places.

With this brief description of the origins of the book completed, we can move on to the more substantive issues covered in this introduction—an overview of the case studies and the criteria for their inclusion, our definition of success, and the main lessons of the case studies.

The 23 case studies included in this volume

The case studies included deal with a range of policing problems which fell into five general problem types: (1) gang violence, (2) violence against women, (3) vulnerable people, (4) disorderly places, and (5) theft, robbery and burglary, a set of problem types much smaller than the 18 general types listed on the website of the Center for Problem-Oriented Policing.²

Most of these case studies had been judged as finalists or winners in the years they were submitted for the Herman Goldstein Award for Excellence in Problem-Oriented Policing and, while the award is open to submissions from any country, most have been submitted by police agencies from the English-speaking countries of the United States, the United Kingdom, Canada, New Zealand, and Australia. In some but not all cases, the authors of these case studies had played some part in the original submission for the Goldstein Award. Where this was not the case, authors chose to write about the project because it illustrated some general themes that they were interested to discuss.

In order to bring some consistency of approach we asked the case study authors to organize their chapters roughly as follows: (1) summarize the project—including how the problem was identified, analyzed, addressed, and assessed (the SARA model)³—drawing from the police agency’s original project report; (2) follow up with the police agency to learn what became of the problem in the years after the project was closed; and (3) discuss any aspect of the project that might be of interest to a wider audience of scholars and practitioners. To assist authors in meeting our charge, we sent them an exemplar case study.

Defining “successful”

We mentioned earlier that several systematic reviews of published evaluations of problem-oriented policing had found it to be effective. Several of the systematic reviews used the Maryland criteria (Sherman et al., 1997; Welsh & Farrington, 2002; Weisburd et al., 2010) to judge effectiveness, but in fact, these criteria are biased against problem-oriented policing. This is because their top scores are reserved for projects that utilize randomized controlled trials, which have very rarely been used in evaluating problem-oriented policing projects; in fact, most such evaluations make use of before-and-after comparisons or simple time series with or without controls. These are deemed weak designs under the Maryland criteria, and while some of these reviews concluded that problem-oriented policing is effective, this was at “promising,” the lowest level of effectiveness.

In fact, unlike treatments to which people can be allocated randomly, such as counseling or enhanced care for juveniles, assigning problem-oriented policing interventions randomly is very difficult. One important reason is that while problem-oriented policing projects are led by police, they are frequently implemented with assistance from other local agencies, which are likely to be even less persuaded of the need for rigorous evaluation than the police might be. They would probably regard evaluation as an academic luxury and would argue that if there were any real doubt that the changes would be effective, why go to all the trouble of making them. They would also generally prefer to focus interventions where they were most needed rather than where dictated by random assignment. Thus, the randomization requirement would multiply many times the usual difficulties of obtaining cooperation and consent to an evaluation.

Even in the rare cases when it might be possible to use random assignment, doing so under the “double-blind” conditions that make these designs so useful in evaluating new drugs would be impossible. (Double-blind studies control for the “halo” effects of a new treatment by ensuring that the clinicians and those treated do not know who has received the drug and who the placebo.) Problem-oriented policing interventions would be much harder to conceal. For example, in a project where improved street lighting was randomly allocated between different neighborhoods, this would be immediately obvious to those who received the improved lighting and to those who did not. Another important reason why random allocation could never serve routinely as the preferred evaluative strategy for problem-oriented policing interventions is the much greater cost of prospectively undertaking random allocation studies compared with retrospective quasi-experimental designs. Testing the effectiveness of problem-oriented policing projects using what are widely regarded as the most rigorous of scientific standards and methods is impractical in all but rare instances. Moreover, some long-held assumptions about what constitutes an appropriately rigorous standard have been challenged of late, the seemingly bedrock principle of “statistical significance” being one. Challengers have called for a more nuanced and gradated interpretation of statistical significance in lieu of a binary one (Amrhein, Greenland & McShane, 2019).

Fortunately, the drops in crime and disorder achieved by problem-oriented policing initiatives are often so precipitous that simple time-series analysis leaves no doubt the interventions have been effective. Ross (2013: p. 129) has called these “cliff edge” effects, (see also Perry et al., 2017) of which there are two main kinds: (1) where an ongoing high rate of crime or disorder suddenly falls after an intervention and (2) those where a specific form of crime that has quickly risen drops precipitously after the intervention. As argued by Nagin and Weisburd (2013), drops of these kinds would require no detailed statistical treatment of the data. In commenting on drafts of the case studies for this volume, we asked authors to provide any data they could that might indicate a precipitous drop in the problems addressed in the case study. In half the case studies, the assessment of impact revealed what can fairly be deemed “cliff-edge” reductions in the incidence or harms caused by the problem.⁴ Whether or not they yielded “cliff-edge drops” in the problem, all of our case studies demonstrated measurable improvements in the targeted problem that were plausibly caused by the problem-oriented policing intervention and to a degree that satisfied key stakeholders in the problem.

We had also asked authors to check in the course of their follow-up inquiries with the responsible police departments whether the same problems had subsequently reverted to their previous levels. Among the case studies included in this volume, there were at least four where this did occur. The most likely reason is that police agencies often fail to institutionalize problem monitoring so that when the key personnel responsible for successfully addressing a problem move on—as they inevitably will—no formal process exists to transfer responsibility for maintaining the conditions that led to initial success. Unfortunately, readdressing a problem once it has again become intolerable is often harder than it would have been to intervene early in its reversion cycle.

Few successes in crime and disorder reduction last forever. The police are seldom in the position of being able to claim that they have discovered and implemented the “once-and-for-all” solution to the problem. Their objectives are far more modest: to bring about noticeable improvements in the problem for a reasonable period in the jurisdiction in which they are being experienced. Changing conditions in communities, including when problem solvers move on to other priorities and diminish their oversight of the initial problem, can result in a resurgence of the problem; displacement of it to other locations, times, and targets; or unintended facilitation of new problems.

The relevant standards of success in highly complex and inherently political environments such as policing do and should differ from thos...