Title: Measuring What Matters Part Two: Developing Measures of What the Police Do. Subject: Information systems, police planning and management, problem-oriented policing, media Author: Thomas V. Brady Series: NIJ Research in Action Published: November 1997 41 pages 71,000 bytes Figures, charts, forms, and tables are not included in this ASCII plain-text file. To view this document in its entirety, download the Adobe Acrobat graphic file available from this Web site or order a print copy from NCJRS at 800-851-3420. ------------------------------------------------------- U.S. Department of Justice Office of Justice Programs National Institute of Justice Research in Action Jeremy Travis, Director November 1997 ------------------------------------------------------- Measuring What Matters Part Two: Developing Measures of What the Police Do A Report From the Policing Research Institute Sponsored by NIJ and the Office of Community Oriented Policing Services When we worked for police departments, we kept a managerial eye on our friends in the private sector. The bottom line told them whether what they did mattered. Economic success or failure was the clear measure of their performance. You do well, you flourish; you perform poorly enough, you close your doors. Police departments don't go out of business; good or bad, they survive. So absent the unequivocal report card that the bottom line in business provides, how do police departments (1) know what they are doing matters, and (2) how do they measure what matters? These are not idle squadroom questions. Policing is a costly, complex enterprise whose practitioners intervene with citizens 24 hours a day on matters ranging from homicide to blaring radios. Depending on your angle of vision, policing is crucial to the aggressive control of crime; a vital partner in maintaining a community's cohesion and sense of well-being; essential to neutralizing the forces of disorder and enhancing the quality of life in neighborhoods; and the vigorous source of innovative solutions to chronic problems, whether juvenile gun violence, for example, or street corner prostitution. Jurisdictions may assign one, two, or all of these purposes to their police departments. Whatever police departments are asked to do, our questions are central to understanding and judging their operations even though we know the answers are sometimes stubbornly difficult to achieve. To repeat, how do police departments (1) know what they are doing matters, and (2) how do they measure what matters? They are questions we brought to Washington when we were appointed to run two Department of Justice agencies that deal with policing. To explore answers, our agencies--the National Institute of Justice (NIJ) and the Office of Community Oriented Policing Services (COPS)--sponsored three meetings in 1995-96. We invited police officials and criminal justice scholars and researchers to take part along with representatives of community organizations and journalists. The meetings, entitled "Measuring What Matters," were held under the auspices of the NIJ/COPS Policing Research Institute. We expected no tangible or final answers to our questions. Experience told us measurable bottom lines in policing are elusive. Indeed, the participants came to no uniform conclusions. There is an immediate tension in the room when you ask criminologists and police chiefs to consider measuring what the police do, observed Carl B. Klockars, professor of criminal justice at the University of Delaware. Klockars said: "Criminologists believe measurement is difficult, complicated, time consuming, and expensive. Inferences from even the best measurements must be made with great caution. Police leaders believe measurement ought to be easy, simple, quick, and cheap. Inferences should be obvious, helpful, and, whenever possible, favorable." The meetings did surface in one arena many of the considerations and ingredients involved in measuring what the police do. Moreover, the final meeting provided useful and novel perspectives and ideas. Some examples: o Police should consider having a Uniform Crime Reports (UCR)-like index measuring disorder as a companion to the FBI's UCR on major felonies. o There are six attributes of a healthy police agency, and these attributes provide grist for measurement. o An important part of police work is best assessed through police-citizen encounters, which have four aspects--fairness, civility, concern, and apparent effort. o To get a different perspective on measuring themselves, the police can view themselves through the prism of business and industry so that they see their work as providing products to customers in a marketplace. A report of the first of the three meetings, held in late 1995, captures the initial, at times outspoken, debate among 45 participants. The report, Measuring What Matters--Part One: Measures of Crime, Fear, and Disorder, is available from NIJ. This report focuses on the second and third meetings when participants concentrated on police accountability to the public and on developing and implementing outcome-based police performance measures. Seven specially commissioned papers were prepared for the second meeting, which is discussed in the section entitled "Perspectives on the Police" beginning on page 11. Another major section, starting on page 12, deals with the police and the news media. A report of the final meeting, held on December 4, 1996, follows. Jeremy Travis and Joseph E. Brann Before becoming Director of the National Institute of Justice, Jeremy Travis was deputy commissioner for legal matters for the New York City Police Department. Joseph E. Brann, Director of the Office of Community Oriented Policing Services, is a 26-year veteran of police service, including 5 years as chief of the Hayward, California, Police Department. ------------------------------------------------------- What Do You Wish to Measure? The main business of the final meeting was the deliberations of four groups drawn from the meeting's participants. Robert H. Langworthy, director, Justice Center, University of Alaska- Anchorage, and Carl B. Klockars, professor of criminal justice at the University of Delaware, prepared them for their task. The day's proceedings were moderated by Francis X. Hartmann, executive director of the Program in Criminal Justice Policy and Management at Harvard University's Kennedy School of Government. Klockars is a maker of checklists and provider of droll advice. Thus he told police chiefs at the meeting: o Everything you measure can and will be held against you. o Measure all those things that can be counted to your credit. o If you cannot get away with measuring only your credits, measure both what can and will be held against you and what may be counted to your credit. He prepared this checklist of "Cheap Measures of What Really Matters:" Measurement o What, exactly, is it that you wish to measure? The most important thing about measurement is to define the problem that you want to measure in the proper way. o Is the problem worth standardizing? If so, as what? o What is gained and what is lost by attaching numbers to it by some rule? o Do you wish to unleash the power of mathematics to describe it? o If so, who will profit from your doing so? o Who will gain access to the power of the numbers? Cheap o Is measurement, relative to the alternatives, inexpensive? o Is it really worth it relative to benefits? o Can measurement be accomplished as a side benefit of doing something else that ought to or has to be done? Is it "free?" Does it really matter? o What do I gain by measuring? o What do I risk by not measuring? o Does it measure success as well as failure? o Can it be made to measure success only? Measure What You Can Control Other advice Klockars offered police chiefs: "Measure what really matters, by which I mean things you can be held accountable for and things that are under your control and about which you can actually do something. One of the difficulties about the police relationship with crime is that although the police are often evaluated by whether crime goes up or crime goes down, they don't have the capacity in many ways to control it. In some areas they can; in others, they can't. So the idea is, mobilize measurements for which you properly can be held accountable and which can count what you do." Like others at the meeting, Klockars urged using data from computer-aided dispatch (CAD) systems as a bountiful source of information for measurement. CAD data provide a valuable "description of what in fact police do day in and day out" and what people want from the police, Klockars said. He also urged the police to measure "because it's under your control to change how well your people deliver service to the people who ask for it. I think all these measures of fear in the community really are pointless. They are so heavily mediated by what people see on the 5 o'clock news that they are not under your control. What is under your control as a police chief is whether officers delivered requested service." (For other viewpoints on the news media and the police, see the section on page 12.) Klockars noted the enduring debate between social scientists and the police about the quality of police statistics. "Police reports of crime are subject to a variety of types of discretion. Citizen discretion in reporting--police don't know about crime generally speaking unless somebody reports it. There's perpetrator efforts at concealing crimes that go on. There's police discretion in how they record crimes. Finally, in a large variety of crimes for which there is often no immediate complaint--gambling, prostitution, corruption, etc.--all police statistics measure is the amount of resources that they place in discovering those things." Yet, he said, "Despite all the criticisms that criminologists have of police statistics, they're absolutely irresistible and the reason is that they are free. Cops generate them in the course of doing the work that they do." Klockars suggested ways to measure police competence, skill, and integrity. His ideas are contained in a detailed paper that is included in a forthcoming NIJ volume of all papers prepared for the three meetings. Langworthy stressed a need to develop measures the police can routinely collect information on and make them meaningful to the people who use them. He said the measures should tend to be of the type Klockars said police leaders prefer--easy, simple, cheap, and from which useful inferences can be made. Langworthy assigned to the four groups their areas for discussion, which he called conceptual domains. o The impact domain is a cluster of items that the police are supposed to affect--crime, fear of crime, and disorder. o The organizational health domain deals with the nature and volume of police business and community support as well as the level of job satisfaction of police employees and their knowledge of their jobs. o The process domain deals with fairness, civility, equitable service, and ethical service, and the community assessment domain deals with police abilities and ethical behavior. Both address the way the police do their work. o The community context domain involves social cohesion, informal social control, and political and social structures. ------------------------------------------------------- Group One Report: A UCR for Disorder Group One, which explored measures for crime, fear of crime, and disorder, suggested the possibility of creating a national index for disorder tied to the Uniform Crime Reports (UCR) the Federal Bureau of Investigation publishes on major felonies. Edward Flynn, chief of the Chelsea, Massachusetts, Police Department, explained: "At the end of the day, when we're talking about measurement, we're inevitably talking about comparisons. Everybody wants to know how they shape up, how they compare. When you tell them about their problem, it's 'compared to what?' "Basically, over the years the conventional wisdom about policing is that 80 to 85 percent of it is not crime related. Yet the only thing we measure and track for the whole country is the 20 percent that is crime related. Eighty percent of our picture is just out there, and we're just kind of dealing with it in a very localized way. "So we thought, 'What if you had a UCR index for disorder?' You try to track it in the aggregate using calls for service, for all their imperfections. If we can impose on our calls for service some basic order of standardization for a noise complaint, a domestic violence complaint, a kids-on-the-corner complaint--if we can impose that basic order--what we would have over time is a measure of the reality of the public's perceptions of the problem [of disorder]. "It's not so much that I need to know exactly how many prostitutes on the corner I have, no. But I need to know how many calls about prostitutes I have because that tells me about the concerns and priorities of that neighborhood. We have all been to the same block associations. Nobody at a block watch ever complains about bank robbery. We all know what they complain about, and then we all go off and deal with it all alone, with no context. "What we're trying to come up with is a way to take the huge amount of police-community work and put it in a context that I can pull out the Uniform Crime Reports and say, 'This is how Chelsea shapes up compared to cities of similar size.' You're not alone from a comparative point of view. From a specific point of view, yes, I need to know what that's telling me about my neighborhoods. "But it's a huge amount of data we do collect, and we thought maybe there's a positive use that can be made of it, that we can come up with half a dozen index crimes, come up with a half dozen key indicators of complaints to police that tell you about the shape of your city. If you monitor the calls per thousand, they probably tell you a heck of a lot more about your neighborhood's or city's health than the larcenies per thousand. "And the thing about trying to hook it into the UCR, that's already fairly cheap because it's built into our costs. It's already got credibility, for all the shortcomings of it, to the broad community, society, and media. If it's UCR-approved, it's real." Mary Ann Wycoff, senior research associate for the Police Executive Research Forum, added, "We all know how disastrous call coding is in trying to capture incidents that come under the heading of disorder. We're hoping that some major departments might consider the idea of creating a UCR-like index for disorders and help to develop strong measures to back it up. The suggestion was made that efforts begin in metropolitan areas so that you bring clusters of departments together to work on them. Perhaps you can then start encouraging many other departments to seriously examine their coding processes and rethink how they collect and store data on disorders. Eventually, this would help everybody with problem solving." Data Can Be Used Against You Jean Johnson, director of programs for Public Agenda, a nonpartisan research organization, presented the overall report of the group's discussions. She said a cautionary "overarching theme" emerging from the group was: "If we're going to measure certain things, you have to really keep in mind how the data are going to be used. Data can and will be used against you not only by the press in a political sense but also it can be used in sort of a departmental rivalry in municipal areas." First, she said, the group worked "on an inventory of indices focusing on quality of life, coming up with items such as truancy, vacant buildings, unemployment, AFDC [Aid to Families With Dependent Children] participation, public housing, whether there were afterschool activities, property values, demographic changes, influx of immigrants to the area. There was a sense that these items, while it's worthwhile to measure them, are--for a lot of police departments--their context. These may be warning signs; they may allow you to compare how you're doing with other cities; but you can't really do that much about them." Next, she said, the group worked on "indices to measure what people are worried about and what the police are reacting to in the category of disorder--calls about the unruly, unsupervised kids, reports of prostitution, reports of dealing drugs, noise complaints, domestic violence, abandoned autos, public drunkenness. The group concluded that most police chiefs need measurements that are quite specific and accurate for them to be useful. "In terms of the source for measurements, we focused on calls for service and expanded that to include the self-initiated action of police officers when they're out on patrol. "There was [also] a feeling that if people call the department, they think that something is very important and they want the police to do something about it, and that's an indicator in and of itself. It's a kind of urgency measurement." Ralph Taylor, a professor of criminal justice at Temple University, said an idea "that emerged was, `Let's see if we can standardize calls for service.' At the front end of that, of course, there were a range of questions about what does that mean. We didn't want to get grandiose, but at one level you could think of a range of validation studies or criterion studies that sort of explored the meaning of these reports. Then one sets up [measures that are] sort of agreed to across jurisdictions. "The other idea is following up on samples of calls for service because the question is not only what are the problems and where are they, but, when the police went out, did they fix them? So there could be callbacks to a sample of folks who called the police to see if things did get fixed." ------------------------------------------------------- Group Two Report: The Six Attributes Of a Healthy Police Organization Group Two dealt with organizational health and developed what it called the six attributes of a healthy police organization. The group's findings were reported by Stephen D. Mastrofski, professor of criminal justice at Michigan State University. 1. "The healthy police organization knows what it wants to accomplish. It has articulated goals. When we say goals, we're not talking about `To serve and protect,' which is obviously too general. We're talking about goals that can be expressed in an operational form. These goals can be appreciated by the people--certainly middle-management supervisors and especially the rank and file--who have to carry them out. The goals can be assessed, meaning that there are measures, indices, of things that are reflective of the goals." 2. "The healthy police organization needs to know its citizens. Are they getting what they want? What they are entitled to? They are not just those who call and complain, who summon the police, or who rattle the cage of the police. Those who don't surface are in the ambit here, residents in a neighborhood, businesses, and so on. Finally, there are those whom we often think of as the objects of police control--the offender, the troublemaker--they, too, are people who need to be considered in terms of their experiences with the police. There are a variety of user surveys that could measure transactions with citizens, for example, periodic citizen surveys of the community. We think it is very important to break them down into units that matter. Oftentimes, the breakdown will be for patrol officers, for example, at the beat level." 3. "The healthy police organization knows its business, the demands that are placed upon it. Calls for service are a cheap and readily available source of information in this regard. The department needs to know why `business' is increasing or decreasing. And knowing more about business wouldn't be limited to relying on calls for service. There are a variety of other ways that business comes to police. Special efforts would have to be made in terms of measuring proactive efforts by officers, particularly trying to capture programmatic efforts. For example, problem-oriented policing requires not just random responses or responses to individual incidents but responses that are planned and coordinated to accomplish some objectives." 4. "The healthy police organization knows what it's doing about the demands of business. It has the ability to monitor resource allocations and officer activities. In terms of community policing, it knows what other agencies and organizations are doing that are pertinent to the problems it's trying to deal with." 5. "A healthy police organization knows its people. Things that would tell us what people get from their jobs, what they are looking for from their jobs, what motivates them about their work . . . what demoralizes them. Knowing these things would help drive decisions about supervision, training, recruitment, job design. . . . The obvious implication in terms of measurement here is conducting surveys which, in this case, are relatively inexpensive because you have a more or less captive audience within the organization." 6. "The healthy police organization feeds back information to people and groups who need to know. To do this, you need to know what they need to know, what they want to know, and how they need to get it. Whether it's neighborhood groups that need to know more about the kind of service that they're getting, whether it's victim groups, or whether it's constituents within the organization. They are all users of information." Measuring Corruption Discussions on organizational health turned to ways of measuring corruption. "There are some measures that trigger a need for additional, intense measurement," according to Michael E. Clark, president of the Citizens Committee for New York City. "One of those is the historic problem in New York--the dilemma of corruption. If corruption complaints go up, that's treated by the media as a sign that things are bad; if corruption complaints go down, that's treated by the media as a sign that things are bad. How to fix that? "One way is by targeted sting operations in cases where you have some indication that there is a problem. If you do that in a random enough way, you're able to pick up a statistically valid sort of measurement of what the level of the problem is. You also seem to be acting to do something about it [corruption] at the same time, which is a way of staying off the horns of that dilemma." William Bratton, former police commissioner of New York City, described the "twin dilemma" of corruption in similar terms. "If it goes down, you're not doing enough; if it goes up, you're not doing enough." When he was commissioner, Bratton said, "We went to professional pollsters who indicated that with an organization with 38,000 police you'd need to do 700 stings each year to have after several years a valid base to measure whether it [corruption] is up or whether it's down. So, on average, the department now every year will do at least 700 stings. This year they will do almost 900 because of some additional targeting they're doing. After a couple of years, you'll then be in a position, hopefully, that if complaints go up, you can argue it's as the result of the work you're doing rather than an actual increase in the problem." Triggers for Inquiry Group Two also explored the uses of data from computer-aided dispatch systems to detect problems in the field. Comparisons made using CAD data can be "triggers for inquiry," as Langworthy put it. Robert Ford, chief of the Port Orange, Florida, Police Department, described "looking at the mix of calls from CAD and comparing it with the mix of outcomes recorded by the officers responding. There are some powerful data there in understanding where you may have potential problems, and there are some very interesting indicators that could be of value for a lot of departments. "For example, domestic violence. Out of every 100 domestic violence calls assigned by CAD, how many of them come back labeled as domestic violence by the officers? You can say a lot about how officers are handling calls just with that information. So you're talking about a series of measures that come from your own data that, while they may not tell you anything immediate, will tell to look into that issue in more detail. Calls should follow a certain pattern. When you see differences from the normal, it gives you a valuable insight into starting to look for potential problems." Bratton said, "The way it's used in New York, you have 76 precincts. For every 100 calls, if, on average, 85 of those calls, after investigation, are classified domestic violence but some precincts are only showing 75 and others showing 110 percent, those ones that are showing 75--you go in and find out what's going on in that precinct that is below the norm. The same thing for arrests. It's a way of detecting a blip that allows you then to focus more attention on what that blip might be telling you." For Langworthy, "Most of what we are talking about here are measures that really are triggers for inquiry. They're not things that give you any answers; they tell you where you need to start looking. So you make comparisons. "There's a crime analyst who's essentially said there is a normal level of crime. When I see a departure from that, then we have something that we have to try to figure out. It doesn't mean that things are bad. It just means that there is something different there now than there was before. "What we can do if we have systems in place is look at the triggers. That holds for virtually all the things that we are talking about. Absentee rates, for instance. You may find out that absentee rates skyrocket one week. That can be because you got Blue Flu [police staying away from work to protest agency policy], or it can be because you got Type-A Flu. You don't know what the reason is by looking at the blip; you only know that you have to go look." Klockars cautioned, "One of the experiences we have, of course, is that agencies, institutions, quickly develop such things [indicators] into quotas. That kind of thing happens in traffic and in all sorts of other areas. The officer out there whose sergeant says, `The average is 4.1 tickets per X amount of times, and you're only doing three. Let's get going'--that's not an indicator any more, that's a quota." Moderator Hartmann asked about the cost of data used to trigger inquiries. Bratton: "How cheap? There's really no cost when you're using information that's being generated in the course of your normal business. The expense comes when you hit a trigger, you've got to go out and look at it. But crime statistical information, you're generating that anyway. The prioritization of calls--domestic violence, for example--for review to learn what officers' actions are is not really a cost. It's data that are already being generated in the normal course of business but using that information for a different purpose." ------------------------------------------------------- Group Three Report: Citizen Satisfaction, Equity, and Ethical Service Group Three dealt with process and community assessment domains focused on three matters-- categories it called citizen satisfaction with police encounters, equity of police service provision, and ethical service. Wesley G. Skogan, a professor of political science at Northwestern University, summed up the discussions: Citizen satisfaction. "An important part of police work--not all, of course--is best assessed through police-citizen encounters." Encounters have four aspects--fairness, civility, concern, and apparent effort. "The fairness and civility dimensions turn out to be the strongest predictors of relative satisfaction with police-initiated encounters with citizens. The concern and effort indicators are predictors of citizen satisfaction with citizen-initiated encounters with the police. In some way, all four of them play out in almost every police-citizen encounter." Ways cited to measure citizens' satisfaction with police encounters range from officers handing out postcard questionnaires to people they come in contact with to letters sent to crime victims inquiring about their treatment to telephone surveys of people who have various contacts with police service. The group stressed the importance of police organizations taking these survey results seriously, "about putting in place an analysis and reporting system that would use these data internally. In Madison, Wisconsin, for example, after they accumulate for a while, officers get postcards back so they can see what people have said anonymously about the contact that they've had. It's important to find ways to use these data as learning and teaching tools as well as assessment tools." Equity. The group dealt with the equity of police service provision in terms of "equality in the distribution of the public good of policing and the public good that comes from effective policing." Matters such as officer deployment-- "how many officers when, where, on what shifts"-- and measures of service delivery such as response and service time could be obtained "quite easily out of existing departmental information systems. With some work, analytic tabulations that would routinely report on the distributional aspects of the delivery of police services and the equality of distribution could easily be done." Disagreements arise "over the standards by which equality should be judged. In Chicago, for example, officers are distributed in a weighted workload formula that takes into account burglary and other kinds of crimes. Critics of this simply want to divide the number of officers in every district by the amount of violent crime and use that as their only standard. When they do that, they come up with a recommendation for a different distribution of officers than the weighted workload formula." Ethical service. This dimension encompasses "corruption both grand and petty, police use of force, the integrity of the crime-reporting process, and, more generally, internal recording and reporting. It overlaps heavily with fairness because racism, hate bias, and other kinds of unfairness in the delivery of services fall in the ethical domain as well as in the service delivery domain. Some kinds of ethical problems manifest themselves in police conduct in public; other kinds of ethical problems principally manifest themselves within the organization, . . . violation of organizational rules." The Need for Clear Standards and Policies Skogan said Group Three "called for clarity about what standards and policies are within a department because you can't begin to measure things until you have clarity of standards and policies. We talked about how normative, situational, and textual many of these kinds of things are. [It's quite] difficult to develop ongoing, easy flowing indicators of them. We noted that many departments don't routinely collect data on even the most elemental of them, including police use of force. And it's been principally through lawsuits that they have been pushed in the direction of doing so. "We ended up endorsing the idea of realistically, carefully documenting the procedures by which a department attempts to ensure compliance with policies regarding ethical issues--documenting not just public complaints but following up what otherwise is basically a black hole, documenting the extent to which they [complaints] are reviewed, the extent to which the various kinds of infractions, or various kinds of exonerations, have been concluded. But principally we think this is a management issue, not so much a measuring issue." The group also discussed "the practicality of survey research as a way of trying to assess the views and experiences of the general public as opposed to people who have been clients or consumers of police services. Many cities find that they can afford to include police questions in omnibus service delivery surveys that are being done by their cities. Of course, we think that that's terrific. "But many police administrators are going to have to live with much more rough-and-ready indicators of what the mass public thinks. We talked about these as a way of putting people's fingers on the pulse of the community: Informal street corner surveys, brief questionnaires administered in shopping malls and strip malls, informal focus groups with activists and informed citizens, and, very interestingly, using the schools as places where vital constituencies of the police--teenage kids--come together on a regular basis. We can easily envision a survey of high school seniors, for example, talking about crime problems, encounters with the police, quality of life, and that can quite easily be monitored because of the inexpensive capacity to do a kind of yearly senior survey in a school setting that would give you an inexpensive handle on a vital and core constituency." ------------------------------------------------------- Group Four Report: Community Context Indices Group Four discussed indices of factors within the community that point to its condition. The group's report was made by Clarence Harmon, former police chief of St. Louis, Missouri, who, subsequent to the final meeting, was elected the city's new mayor. Among the factors the group examined were: Demography. "Family structure and economic structure indices could be measured in terms of what outcomes we might look at to tell us whether we were improving or not in terms of services." Organizational infrastructure. "Whether community organizations had the capacity to effectively function is an indicator of progress or lack of progress. The number of not-for-profit organizations--whether they were holding meetings- -and block parties being held in communities are positive indicators. Permits for parades and other functions are positive indicators that the community is holding up and that we had made some progress in terms of our service delivery and a community-based effort." Informal social controls. Graffiti is one index. "In California there's a graffiti hotline that a number of departments operate as a mechanism for cleaning it up. Having such a mechanism would be an indicator of whether or not that phenomenon was growing. If you earmarked efforts to reduce graffiti, and it declined in terms of the reports of the hotline, it might be a good index that efforts were successful." Domestic violence is "an indicator of how effective [police] personnel can be in dealing with neighborhood issues, but only, I would argue, to a small degree. There are a lot of other issues that impact domestic violence, not all of which, in my opinion, are easily understood or easily affected by the police--poverty being among them. "Truancy and school attendance, teenage pregnancy, and repeat calls of single incidents are measures of the wellness of a particular community." Indices of social cohesion. "Resident turnover as measured by school enrollment, school turnover in terms of kids entering and leaving, liquor outlet densities, are indices of social cohesion or the lack thereof." Critical incidents. One index is "emergency room admissions. One of the indicators we're looking at in St. Louis is the impact of having taken a number of officers out of drug enforcement and placed them in patrol. One of the indicators that shifting officers has been a problem is an increase in the number of admissions in emergency rooms for heroin addiction and cocaine problems. We're not necessarily seeing an increase in the arrests for those same offenses." ------------------------------------------------------- The Summarizing Session: Policing Through the Prism of Business and Industry An unconventional way of exploring policing and ways of measuring it is to employ the understandings and language of business and industry. Jack Greene, a professor of criminal justice at Temple University, used this approach in the summarizing session of the final meeting. "We are to assist police departments in a process of conceptualizing what matters, identifying how to measure it, and implementing [those measures] in several places with the hope of improving what I would say is the industry," according to Greene. "We're looking at where this industry is going and where individual firms in the marketplace are." The following are excerpts from Greene's summation that used, as he said, "a lot of business jargon." Market dynamics. For police departments, market dynamics are "community context data--information about shifts in population bases, changes in school-age population, changes in elderly population, shifts in welfare policy, level of teenage pregnancy, illiteracy." These data help to indicate "where the products that the police produce are likely to have a market and an impact and an attachment." They typically are available from Federal census information, local universities, and municipal research departments. "Unfortunately, many police agencies have been reluctant to get into a marketplace analysis; more often, police are driven by market dynamics rather than either trying to predict or affect them." Market segmentation. "People don't use the police the same way, and lots of people use the police for different purposes. So the market is already segmented. It's segmented by age; by the location of the community in which people reside; between residential communities and communities that are defined on the basis of retail trade or industry or other factors. So market segmentation gives us a chance to look at subpopulations. It begins to say, `Let's take our marketplace, divide it up into the various kinds of users, and have a better understanding of what the use patterns are, what product identities these various constituents have, and how they use and evaluate services.' It's not passive any longer. It's trying, in effect as a major corporation would, to go out and understand the market dynamics and position products to fit the needs of those client groupings. The data come from several places, not the least of which are computer-aided dispatch data, but from other information as well, perhaps segmented surveys by different kinds of users." Market demand. "How do we take market demand for certain kinds of products and turn that into something usable to the client group that will consume it? All kinds of things are at stake here. How is the service produced? I hear a lot about interactions. How do police officers interact with the public? What kinds of services does the public expect to get? What kind of services do they get? Are they the same, or are they divergent, and by how much? Clearly interactional data--both observational as well as self-reported on the part of both the public and the police--can be used" in these matters. Work force. "We need to know something about the work force. Very clearly the work force is acutely at stake in a labor-intensive industry such as policing. If we don't pay attention to all of the nuances of organizational life, we don't really know very much about why we produce better or less better interactions out there. There is a huge literature and very well rounded instrumentation in the management and measurement of workforce understandings. Do I know my boss? Do we get along consensually? Do I believe in the rules? Do I feel like I'm part of the process? There are hundreds of standard instruments that can very easily and fairly unobtrusively measure and monitor [the work force] on a routine basis to get some sense for the mental and social and communal health of things within the police agency--that is, the production force that's there." Product line. "We sometimes have a one-size-fits-all product with variations on a theme: 'I'm going to arrest you, I'm not going to arrest you, or I'm going to come back and arrest you.' It seems to me that there's a wider range of products than that. Certainly, problem-oriented policing and other interventions have begun to broaden the range of products that allow us to both segment the market at the macro level but also segment the product and line those products up with the markets we are trying to produce some response in." Systems rates. "Systems rates are the normal rhythm of the agency and the production of whatever it is that it's producing, although in justice issues I'm never sure quite what normal represents and whether a high rate of something-- even though it's a constantly high rate--is something that we want to achieve. But there is the idea of systems-rates regularity. Monitoring of all kinds of workflow processes within the organization, as well as the mental health of the workforce, is a way of establishing some baseline information. That baseline information can then be fed back into questions of quality assurance that would get to some of the management issues that I heard about in the integrity question." Cost. "One piece that I find always fascinating in major police agencies is cost. The unit cost of a product becomes extremely important. We just divide the workforce by the number of calls for service and throw in the light bill and somehow that produces a way to define costs. But there are all kinds of costs, some of which are intended, some unintended, and some of which are opportunity costs--alternatives foregone because we invested in other things." Symbolic component of a product. "There's a part of the product that is very difficult to get a handle on, but it's very necessary. It's the symbolic component of the product. I've heard a lot about tangible calls for police service, about crimes reported and not reported, and about the monitoring of whether we can and do make distinctions in domestic violence situations. I think it's very important that we keep our pulse on product identity, the symbolic component, the people attached to that [the product] because it defines a user base and support that will ultimately affect the marketplace." Product improvement. "There is a need to have a system in place that will implement new methods, evaluate those new methods, and use that information to reinform both what can be produced and what we will market to the marketplace as the products of the police. Mostly all the problem-oriented policing (POP) efforts and a lot of the work that has been done in situational crime prevention and community crime prevention are about research and development activity within policing that is reshaping the tools--that is, using that information to reshape interventions-- to bring in different responses besides the arrest, not-arrest, threaten-to-arrest response. Those kinds of things are investments in the long-term R&D [research and development] of this industry called policing." Test marketing. "We need to know about product quality. There is test marketing going on here. When a POP intervention goes on in a particular community and there is an assessment of it, that is the equivalent of test marketing a particular idea. The presumption is that if you do that in enough places with similar enough kinds of circumstances, then we may actually create a new technology to intervene in those [relevant] kinds of situations. So it seems to me that evaluative research, which was off the board, should come back onto the board if [for] nothing else than to inform the research and development curve of this process." Product information. "There's lots of information that we don't know about because we don't monitor or collect a lot of information as systematically as we would like. You go into these places [police departments], you talk to one unit, and they're doing something that the rest of the department has no idea is actually going on. So it seems to me that knowing more about the product and routinizing that product internally through information that captures effort, workload, time spent, and cost is very important. Virtually all CAD systems with some reprogramming can capture most of that, particularly time spent in activities in multiple units. That's not to undercut the importance of problem solving, however. If you pursue the CAD response independently, then problem solving begins to trickle away from that process. That's why I say problem solving is more in the research and development category, which is a kind of new, experimental way of understanding new tools." Industry norm. "I know that this sometimes is not the popular view, but it is important to compare individual, discrete departments against some industry norm. It will be very important to establish what that industry norm is. That industry norm will vary by size of the firm, by regionality of the firm, by resources available to the firm, by command of marketplace, by all of those things that are reasonable for differentiating these agencies. Simply to look at 17,000 places [police agencies] independently of one another just begs the question. A lot of the problems that [police departments have] are not really that different from one place to another despite the claim to uniqueness that one is first met with at your chief's office." Greene's remarks drew a variety of responses. The Public Produces Safety Michael Clark said he had "no problem with the industry analogy and no problem with market-oriented management." But, he added, "You can't just view the public as consumers, customers, recipients, clients. They are also producers. They are people who produce safety in their buildings, on their blocks, in their neighborhoods, every day. When the police learn how to use that energy and catch that wave, they find their job is made much easier. Sometimes that's hard to do because they don't know too much about how communities are organized, but the reality is that when they catch that wave and when they work with it, they take a helluva lot longer ride than when they go out and try to just sort of plunge around thinking that the community has thousands of disparate individuals." Of Justice, Equity, and Safety Warren Friedman, executive director of the Chicago Alliance for Neighborhood Safety, said, "I think I have a little more trouble than Mike does with the industrial market analogy. . . . If I understood you, Jack, the symbolic things were justice, equity, and safety. To me, that's the product. It's very hard to make something like justice, equity, and safety similar to anything a business produces. It is just absolutely different. A business ultimately produces profit. Justice, equity, and safety--if we're not getting departments of government that help to produce them--it seems to me we have a problem." Neighborhoods Perform in Civic and Public Ways David Duffee, professor of criminal justice at the University at Albany, said he found "the market analogies quite useful for some purposes. But the police have to be aware of the extent to which neighborhoods are capable of forming as constituencies; that they perform in civic and public ways. The police have to operate in a way-- and I think one can measure this--that is sensitive to the ability of neighborhoods to act that way. They have to be sensitive to the way in which neighborhoods are organized and to some extent will want to make sure that they don't interfere with that organization and facilitate it to the extent they can. Not that they should be in charge of organizing neighborhoods. But they need to work with the constituency groups that are there." A Public Agency Can't Be a Private Agency Johnnie Johnson, chief of the Birmingham, Alabama, Police Department, commented: "I want someone to tell me how to be a better police agency within the context of my environment. I am not so all fire sure that I want to be a private business. When you look at most police departments as being very small, the conversation that we have going on here is way up above their level of understanding. You're not going to help them there. Somewhere between where we are and where they are, we have to find a place where it's going to be meaningful and helpful to the people who we're trying to affect. "There are some successful businesses out there that we can copy some things out of. However, we can't make a government agency, a public agency, into a private agency. There are too many differences there to make that work. "Marketplace? Product? What about my own words? Why can't I use the words that I use now as a police officer? Why do I have to change my terminology to understand how to progress? I am not saying there shouldn't be some systemic changes, but do I have to completely be reborn in order to be effective?" ------------------------------------------------------- Perspectives on the Police There are several perspectives from which to view and measure the police. One is through the lens of public opinion. "The police enjoy a robust vote of confidence from most of the public," according to Public Agenda, a nonpartisan research organization. "But support for law enforcement has a fault line. Far too many black Americans are disaffected and suspicious." A conventional way of looking at the police is as an integral part--along with prosecutors, courts, and corrections--of the criminal justice system. In this view, police are a kind of intake funnel in the processing of criminal cases. But from a different angle of vision, the police can be seen as an agency of municipal government committed to community problem solving, in the formulation of Mark H. Moore, a professor at Harvard University's Kennedy School of Government. A third way to look at the police--particularly in this era of community policing--is through the eyes of community organizers. "One of the precepts that should guide police work is to do things in such a way that the community does for itself as much as possible and thus develops the habits and skills of doing," said Warren Friedman, executive director of the Chicago Alliance for Neighborhood Safety, and Michael Clark, president of the Citizens Committee for New York City. The police also have perspectives on themselves and their constituents. According to Johnnie Johnson, Jr., chief of the Birmingham, Alabama, Police Department, "Many people belong to internal and external communities in their interactions with law enforcement. These overlapping boundaries include religion, social class, gender, ethnic status, sexual preference, civic groups, and education levels. None of these are our `bosses,' but they all have an impact on the way we do our jobs." Chief Johnson's assessment was contained in a paper written with the assistance of Chief Robert Ford of the Port Orange, Florida, Police Department and Chief Dennis Nowicki of the Charlotte-Mecklenburg, North Carolina, Police Department. Another perspective comes from scholars who examined the expectations of various police constituencies and expectations that the police have of these constituencies. David Duffee, professor of criminal justice at the University at Albany, wrote that the available information "suggests that police spend more time creating constituency support for their own organizational needs in the general city population than they do in facilitating constituency behavior on behalf of neighborhoods." Stuart Scheingold, professor of political science at the University of Washington, wrote that "it is inappropriate to assume that the fight against crime will bring Americans together and that a reduction in the crime rate is, therefore, a sufficient gauge of successful policing." The media provide another perspective on the police. Most Americans "learn about crime and the police from entertainment shows . . ., from the police news roundup in their local newspaper, and from the lead stories on the local TV station," according to Aric Press, a senior writer for Newsweek, and Andrew Benson, a former reporter for the Cleveland Plain Dealer. Their paper is excerpted in a section entitled "The Police and the News Media" (page 12). The seven papers were prepared for the second of the three "Measuring What Matters" meetings held on May 13, 1996. They provided the framework for daylong discussions focused on expectations of the police by the general public, the media, community organizations, local government, and other police constituencies. Some key points are excerpted here. The Fault Line in Support of the Police Relying on public opinion research, Jean Johnson, director of programs for the New York City-based Public Agenda, reported in her paper that: "In a decade when many Americans seem to think that "government" can do no right, law enforcement is viewed as an essential public service. In a decade when most Americans are skeptical about government and many other social institutions as well, the police enjoy a robust vote of confidence from most of the public. But support for law enforcement has a fault line. Far too many black Americans are disaffected and suspicious. They are not confident that the police will be fair. They are not confident that the police will be professional. They are not confident that the police will "protect and serve." And while the personal encounters most whites have with police officers may be positive, white Americans have now witnessed two highly publicized and graphic instances of police behavior (the Rodney King beating and the attitudes associated with former Los Angeles police officer Mark Fuhrman) which, in their view, is entirely unacceptable. They regard these incidents as exceptions, but not exceptions to be glossed over as "the cost of doing business." The Police as an Agency of Municipal Government Mark Moore's paper called on the reader to "imagine all the different ways in which the police might contribute to the overall performance of municipal government" and not only to "the just and effective operations of the criminal justice system." The police, thereby, could become "the agency of city government responsible for controlling and preventing crime, reducing fear, and supporting both public and private agencies in their important work." But this augmented role would change significantly "our current understanding of the mission, goals, and objectives of the police." New capabilities would be needed to measure this enlarged role. "One such capability is a large, continuing capacity to survey citizens," Moore wrote. For example, there should be "a general population survey that is designed to capture information about (1) criminal victimization; (2) reasons for not reporting crimes to the police; (3) general attitudes toward the police; (4) levels of fear; and (5) types of self-defense they rely on to supplement the protection they get from the police." Moore also suggested a customer survey administered to a sample of persons who call the police and even a survey of persons stopped or arrested by the police. A second capability "the police must develop is a continuing capacity to evaluate their own proactive problem-solving efforts." Communities and the Police Warren Friedman and Michael Clark offered several guiding observations about communities and the police. "The importance of organizing and organization cannot be overemphasized. The community needs to be organized because it is the most effective way to work with a highly organized partner like the police and against often well-organized and frightening adversaries. [W]ithin a version of community policing that assumes its distinctive strategy is a problem-solving partnership, there are five defining characteristics: It is (1) a coordinated working relationship (partnership) between (2) organized members of the community, the police, other government agencies, and other institutions to (3) solve chronic crime and disorder problems at specific locations and to (4) secure reductions and improvements by (5) increasing the community's capacity to contest the recurrence of the problems. One of the precepts that should guide police work is to do things in such a way that the community does for itself as much as possible and thus develops the habits and skills of doing." The Way Police and Their Organizations Operate Chief Johnnie Johnson's paper discussed the public's lack of understanding of the police's role. "The public often does not understand, and perhaps does not want to understand, the way police and their organizations operate. Police generally encounter people at their worst, not their best. They are called to family fights, not picnics. They see mostly the dark side of human nature. Someone has to deal with the blood, the hurt children, and the human anguish that no one wants to face, and it is usually the police. Certainly there is often mistrust and a "us-them" attitude among police, which leads to a chasm in police-community relationships." ------------------------------------------------------- The Police and the News Media Two enduring adversaries, the police and news media, were the subject of recurring comment in the second and third Measuring What Matters sessions. Topics included the power of the news media, how to deal with the media, and "How the News Media Help Shape the Public's View of the Police." The last topic was the title of a paper contributed to the second session by Aric Press of Newsweek and Andrew Benson, formerly a reporter for the Cleveland Plain Dealer. It All Boils Down to the Media "No matter what we come up with measurements, it all boils down to what the media determines to be important," Dennis Nowicki, chief of the Charlotte-Mecklenburg, North Carolina, Police Department, told the final session. "Somehow we ought to be addressing the issue of the media in our lives and in how it affects people's understanding or feeling about what matters. You could design these constructive measurements and making outstanding progress in the right direction over time and still be perceived by your community, by your elected officials, and by others as a terrible organization because of the way the media portrays what's going on in your community, in the way the media portrays your organization." Bring Them in for Briefings "If you want the media to do something other than 'if it bleeds, it leads,' which is what they do now, you have got to bring them for background briefings; you have got to start talking to them about the data sets that you're developing and the directions you're going and the kind of work you're doing, and you're not going to see a fast payoff," according to Michael Clark, president of the Citizens Committee for New York City. "Investments by definition take time. You're going to see over time a qualitative shift, under the best of circumstances anyway, in the kind of reporting you get. You're not going to control the media, but you do in fact influence their thinking if you work with them on a regular basis." Police Data Is News Andrew Benson, who is now president of the New Ohio Institute in Toledo, Ohio, said, "Chiefs should remember that they have a lot of power in that they have at their control data. Data is news. How you shape the data--honestly, obviously- -can have enormous impact. Right now, you're sending out [possible] failures--crime statistics, response times easily available on CAD. If you say [other things are] important for these reasons, and I'm releasing them, the media will pick it up." He also advised, "Build partnerships with the community, build alliances; then you have strength, you have support, when things are happening. The media will listen to you." Cops and Reporters--Us and Them The paper by Aric Press and Andrew Benson examined in succinct terms police-news media relations and the perception and reality of crime. "The police want "good" press. By that they mean favorable reports that emphasize bravery in the field and wisdom at headquarters. Good press is also the absence of bad press. Bad press in this context describes abuse, corruption, and other mistakes. Cops, like reporters, see the world as divided into two parts--Us and Them. Rather than leading to a mature understanding of each other's roles, these attitudes can lead to hostility. [Through the efforts of the news media] what is the impression left on the public of the police? It is an agency that announces crimes, makes arrests, has a few ideas, struggles with labor-management issues, suffers from some corruption, employs a few brutal officers who may or may not live within the jurisdiction, and appears to be led by a succession of well-meaning administrators who do not seem to last very long." The Reality and Perception of Crime For most Americans, the reality of crime is what they see on television or at the movies and what they read in the newspaper or in a magazine. An overwhelming majority of citizens report they have not been a crime victim in the past year nor do they know anyone who has been a crime victim. . . . The research seems clear that the news media have pervasive, unintended, and unpredictable influences on public opinion. . . . The evidence also strongly suggests that the steady stream of crime news from the media affects the public, such as they are more fearful about the risks of crime than need be and are more likely to demand punitive criminal justice to control crime. . . . The demand by the public for a specific response to crime is likely to lead policymakers to heed the public or, at the very least, to make it more difficult for policymakers to get support for responses that are counter to public opinion.... As a way to address the negative effects of news media accounts, criminologists and journalists have called for more context in crime stories. . . . By tying in the trends, patterns, and causes of crimes, the public would get a better picture of what crime is occurring, where it is occurring, and how often. That gives them information by which they can make informed decisions about their personal safety. This should lead criminologists and police administrators toward providing more of the statistics and research data to the public through the news media. . . . It makes sense that crime news be accompanied by statistical data or inferences from administrators that bring context and order to the seemingly unconnected series of crime and violence emanating from television and newspapers." ------------------------------------------------------- A Police Chief on Police Data "I should warn you that while 80 percent of the departments have computer-assisted dispatch, only about 50 percent of them work. "The state of police data out there is horrible, particularly in terms of their ability to manipulate it. One of the things to think seriously about for a lot of departments is assisting them in improving their data analysis capability. The second point is going to be contrary to what I just said . . . It's amazing how robust information is in the system. If you try and collect it a little, you're going to get a good feel for what the public wants. They're not quiet about it. You're going to get a good sense of what neighborhoods have problems and the like. "I think we can get away with a lot lower level of measurement and still have dramatic impacts. I don't think it has to be that perfect. We can do a helluva lot better than we're doing now because the information there, it's retrievable, it may not be as elegant as we could want, but even nonelegant data will work a lot for us." --Chief Robert Ford, Port Orange, Florida, Police Department ------------------------------------------------------- Screwing Up Your Context ". . . [A]bout the things the police can't do anything about but need to know about. You need to be aware of policies that are working against your police goals. Obviously, the government is responsible for a number of those. If you're trying to clean up and put order back into a public housing project, the fact that the banking industry is working against you is no help whatsoever. Somehow in terms of knowing context, you have to know about other policies that are screwing up your context." --David E. Duffee, professor of criminal justice at the University at Albany. ------------------------------------------------------- Consumer Measures of Police Effectiveness There is "a growing need for independent consumer measures of the effectiveness of the police. Most of the measurement talked about here assumes that the people analyzing the measurements are managers. Being a manager, I can appreciate that. That is a valuable and important task. But an equally important task is for a kind of Consumer Reports version of that same information that monitors the quality of police service and gives the public an opportunity to develop a learning curve of its own about what is good policing. "We have done a rotten job of marketing community policing both to the public and to police officers in the sense of a clear, simple, day-to-day notion of what to expect from the police and what to expect from citizens to make it work. "On the community assessment side, it's very important that we be careful not to jump whole hog into a kind of pathology model. We are ourselves increasingly looking at community assets and at community asset assessment along with problem assessment. It's quite important that we not in a subliminal way transmit a message that communities with lots of crime problems are somehow inherently pathological." --Michael Clark, president of the Citizens Committee for New York City ------------------------------------------------------- Fair Warning Notice "Measurement is an activity so fraught with mind- and soul-wrenching difficulties that only grossly ignorant beginning students and the least capable or least virtuous of social scientists engage in it with good humor. "Fair warning is thus in order to any police practitioner who is approached by a quantitative criminologist with a smile on his face: "Listen very carefully, keeping one hand on your wallet and the other on your gun." --Criminologist Carl B. Klockars, in discussing measuring what matters, issued what he called a "fair warning notice." ------------------------------------------------------- Papers prepared for this and the other Policing Research Institute meetings are scheduled to be published in book-length form by NIJ. ------------------------------------------------------- Selected NIJ Publications About Policing Listed below are some recent NIJ publications related to issues of policing. These products are free and can be obtained from the National Criminal Justice Reference Service: telephone 800- 851-3420, e-mail askncjrs@ncjrs.org, or write NCJRS, P.O. Box 6000, Rockville, MD 20849-6000. These documents can be viewed online at the Justice Information Center World Wide Web site at http://www.ncjrs.org. Please note that when free publications are out of stock, they are available as photocopies or through interlibrary loan. Brady, Thomas V., Measuring What Matters: Part One: Measures of Crime, Fear, and Disorder, Research in Action, a joint publication of the National Institute of Justice and the Office of Community Oriented Policing Services, 1996, NCJ 162205. Building Knowledge About Crime and Justice: The 1997 Research Prospectus of the National Institute of Justice, 1997, NCJ 163708. Criminal Justice Research Under the Crime Act-- 1995 to 1996, Research Report, 1997, NCJ 166142. Dunworth, Terence, Peter Haynes, and Aaron J. Saiger, National Assessment of the Byrne Formula Grant Program, Research in Brief, 1997, NCJ 162203. Feins, Judith D., Joel C. Epstein, and Rebecca Widom, Solving Crime Problems in Residential Neighborhoods: Comprehensive Changes in Design, Management, and Use, Issues and Practices, 1997, NCJ 164488. Finn, Peter, and Julie Esselman Tomz, Developing a Law Enforcement Stress Program for Officers and Their Families, Issues and Practices, 1997, NCJ 163175. Fleissner, Dan, and Fred Heinzelmann, Crime Prevention Through Environmental Design and Community Policing, Research in Action, 1996, NCJ 157308. Gaffigan, Stephen J., and Phyllis P. McDonald, Police Integrity: Public Service With Honor, Research Report, a joint publication of the National Institute of Justice and the Office of Community Oriented Policing Services, 1997, NCJ 163811. Garner, Joel, John Buchanan, Tom Schade, and John Hepburn, Understanding the Use of Force By and Against the Police, Research in Brief, 1996, NCJ 158614. Lively, G. Martin, and Judy A. Reardon, Justice on the Net: The National Institute of Justice Promotes Internet Services, Research in Action, 1996, NCJ 158838. Mastrofski, Stephen D., et al., Law Enforcement in a Time of Community Policing, Research Preview, 1996, FS 000149. Mills, Gregory, COMmunity-Backed Anti-drug Tax: COMBAT in Jackson County, Missouri, Program Focus, 1996, NCJ 160937. Policing in Emerging Democracies: Workshop Papers and Highlights, Research Report, a joint publication of the National Institute of Justice and the Bureau of International Narcotics and Law Enforcement Affairs, 1997, NCJ 167024. Rich, Thomas F., The Chicago Police Department's Information Collection for Automated Mapping (ICAM) Program, Program Focus, 1996, NCJ 160764. Roehl, Janice A., et al., National Process Evaluation of Operation Weed and Seed, Research in Brief, 1996, NCJ 161624. Sadd, Susan, and Randolph M. Grinc, Implementation Challenges in Community Policing: Innovative Neighborhood-Oriented Policing in Eight Cities, Research in Brief, 1996, NCJ 157932. ------------------------------------------------------- This publication was prepared by Thomas V. Brady who is a Washington, D.C.-based writer and editor specializing in criminal and civil justice matters. Findings and conclusions reported here are those of the author and do not necessarily reflect the official position or policies of the U.S. Department of Justice. The National Institute of Justice is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. NCJ 167255 -------------------------------------------------------