Measuring What Matters Part One: Measures of Crime, Fear, and Disorder . Series: NIJ Research in Action Published: December 1996 44 pages 76,895 bytes U.S. Department of Justice Office of Justice Programs National Institute of Justice National Institute of Justice Research in Action Jeremy Travis, Director December 1996 ------------------------------ Measuring What Matters Part One: Measures of Crime, Fear, and Disorder by Thomas V. Brady America develops yardsticks for its major concerns. We measure everything from the productivity of factory workers to rates of infectious illness to the endurance of shortstops. When it comes to public safety, the measurement familiar to most people is whether reported crime is up or down. But other ways exist to gauge crime and its effects as well as to measure the performance of police who share responsibility for public safety. They are the products of research, development, and practice that began 30 years ago and are still evolving. This report listens in on a candid, wide-ranging discussion of crime, policing, and these measures. We hear the distinct voices and opinions of police chiefs, researchers and scholars, and leaders of community organizations. The value of the discussion is plain for those who see the need to assess both crime and police performance in accurate, useful, and innovative ways. Crime exacts a tremendous toll on communities. The police are one of the most expensive parts of local government. They are also most communities' only 24-hour-a-day service agency and the face of local government to many citizens. And, as anyone familiar with the term community policing may know, police agencies are changing the way they operate. Questions arise. For example: What should the police be doing? How do we know when they are doing it? Many people have an official interest in measurements of crime and the police. They include mayors, city council members, county supervisors, and local government managers as well as policymakers and other officials at the State and Federal levels. In addition, community and neighborhood leaders, businesspeople, and plenty of other citizens may want to know about these measurements. Crime is often one of their major concerns. The police are key to resolving those concerns. Finally, students of government and criminal justice and the police rank and file have an obvious interest in the subject. In late November 1995, 45 police officials, criminal justice scholars and researchers, and community organization representatives met for a day of discussion on the topic "Measuring What Matters." The meeting's purpose, according to Jeremy Travis, Director of the National Institute of Justice (NIJ), was "a critical examination of the relationship between police performance and indicators such as crime, fear, disorder, and citizen satisfaction with police service in the context of changing philosophies of policing." The meeting was the inaugural session of NIJ's Policing Research Institute. It was held in cooperation with the Office of Community Oriented Policing Services (COPS), another agency of the U.S. Department of Justice. "We need to look at and evaluate ourselves and our organizations and what we are trying to do in a different light," Joseph E. Brann, a former police chief and now director of the COPS Office, told the meeting. Two more meetings related to "Measuring What Matters" are scheduled for 1996. The second meeting will explore what it means as a police agency to be customer oriented and accountable to the public. The final meeting will address the organizational issues involved in implementing effective outcome-based police performance measures. Answers From Different Worlds Two questions formed the basis of the day's discussion: 1. How do we measure the amount of crime, disorder, and fear and their effects on the quality of community life? 2. Should we expect police activities to impact on measures of crime, disorder, and fear, and how will we know? The answers came, in the words of one participant, from "different worlds." Police leaders "see measurement as a device for improving accountability in performance," said Mark H. Moore, a professor at Harvard University's Kennedy School of Government. "They need to understand [the] what and how of these measures so they can survive." In contrast, researchers and academics are interested "in the accuracy of measures.... They are interested in the accuracy with which attribution can be made of effects to causes," Moore continued. So police and academics seek different things, he said. "They're just different worlds that we're living in." The different worlds were interpreted by a former police chief who is now an academic. "There is an action-orientation on [the part of] a growing number of police chiefs.... The police chiefs are saying that they are not going to wait for the thoughtful research that I hear talked about at the table today," said Gerald Williams, a former police chief of Aurora, Colorado, who is now director of the Bill Blackwood Law Enforcement Management Institute at Sam Houston State University. "Rather, they're going to try something. If it works, they'll continue to do that. If it doesn't work, they'll try something else." But to say that police and researchers come from different worlds doesn't mean that they lined up as polar opposites in the meeting. There were shared observations on topics such as communities, measuring the process of policing, and citizens' expectations of the police. On communities, for example, a police chief, Robert Ford of Port Orange, Florida, said, "The diversity within communities is phenomenal. Our belief now is that every time we go to solve a problem, we create another problem for ourselves from another group." And a scholar, Peter K. Manning of Michigan State University, talked about "aggregated measures which do not take into account political, cultural, social realities.... Unless and until measurement is based on some sense of community boundaries -- whether it's political or moral boundaries within that group of people who live there and interact -- it doesn't make any sense because you're collapsing apples and oranges, horses, mules, and so on." Discussion of the second question focused in considerable part on the view of William R. Bratton, then police commissioner of New York City, that management changes and goal setting in his department were primary catalysts in the steep decline in the city's crime rate during the previous 2 years. Not all police chiefs at the meeting fully accepted Bratton's claim; not all researchers rejected it. "All the things that we [not just the police but the community, other agencies of government, the business community, and neighborhoods] do are at play in crime reduction," said Dennis E. Nowicki, chief of the Charlotte-Mecklenburg, North Carolina, Police Department. According to George L. Kelling, professor of criminal justice at Northeastern University, "Something very powerful is happening in New York City and in some other cities. My own interpretation...might be the same as Bill's...." Readers will find plain speaking in the report. For example, Chief Tom Koby of the Boulder, Colorado, Police Department said he attended the meeting "with a healthy skepticism, quite frankly, about what researchers and academics can provide us. I don't see a lot." And Wesley G. Skogan, a political science professor at Northwestern University, wrote in a paper for the meeting, "Commissioner Bratton has been quoted in reputable sources to the effect that his `bottom line' is crime reduction. This is an unfortunate view; for example, reforms that managed to substantially heal the breech between the races over police conduct might be a more significant accomplishment." Readers will also find an occasionally wry perspective by participants about their professions' efforts. "Much of our difficulty in measurement stems from our effort to constantly use global, all-encompassing terms -- the trinity of crime, disorder, and fear -- which for me has no greater meaning than disease, illness, and the associated anxieties," said Herman Goldstein, a law professor at the University of Wisconsin. "People in the community [have expectations of the police] that aren't very real, and a lot of the ideas we have about people in the community...aren't very real either," said Chief Darrel Stephens of the St. Petersburg, Florida, Police Department. Some of the most useful comments and observations stand alone as sidebars to the main text. In both, we hear the views of meeting participants in their own words. How Do We Measure the Amount of Crime, Disorder, and Fear and Their Effects on the Quality of Community Life? Three papers -- by Wesley G. Skogan, a political science professor at Northwestern University; by Chief Darrel W. Stephens, St. Petersburg, Florida, Police Department; and by Ralph Taylor, a professor of criminal justice at Temple University -- were prepared as background for the discussion of the first question. Each author gave a brief presentation before the start of a general discussion. Skogan distinguished between what he called high-tech evaluations of police performance and initiatives and "monitoring and assessing, the kind of routine evaluation that intelligent and self-directed police departments need to build into their operations." The latter need to be timely, completed on schedule, inexpensive, and done with existing personnel. High-tech evaluations, however, are not routine. They "are done patiently and very strategically [and] are both extensive and pretty expensive to mount." According to Skogan, "Both high-tech and routine self-monitoring evaluations share two concerns: first, measuring what matters, and second, measuring in a way that matters, in a way that we can draw inferences about things which are defensible and demonstrable." According to Skogan, in both types of evaluations, the people involved -- the evaluators -- must start with the "logic model of the program...which is really critical. Logic models have four components -- intervention, context, mechanism, and outcome. Without specifying exactly the logic model of a program, you're never going to get very far in even describing it, much less evaluating it. "The intervention has to do with what you're doing, the level of effort. "The context has to do with the circumstances and the surroundings in which the intervention is being put in place. "The mechanism is how exactly the program is suppose to affect the outcome. If it is an ombudsman policing experiment...how exactly is it that the thing will mobilize services, and what kind of services can they mobilize? How exactly is it that the effort can be sustained during the hours and days that officers are on patrol? "The outcomes are what the anticipated impacts of the program are, so those anticipated impacts have to be things that are carefully tuned to be responsive to the intervention in the context." Thus, Skogan said, in both self-monitoring and high-tech evaluations, "We need to be concerned about the logic -- the inference -- that it was the program that made a difference." He said that both types of evaluations are interested in specific outcomes but very rarely in "macro things like burglary or homicide." For example, Skogan proposed, if the outcome measure is burglary, "Are we talking about commercial or residential? If street crime, day or night?" He added: "Figuring out exactly what the fine-grain micromeasures should be so that they can be responsive to the logic model" is one of the things that is important for evaluators and police practitioners to negotiate. Stephens discussed the Uniform Crime Reports (UCR), city-by-city figures on major felonies that are collected and published by the Federal Bureau of Investigation. "Despite all the limitations of the UCR, it's our only real tool that...police departments have to talk about crime that's happening in the community.... The reality is that periodically we get held to account for those numbers by media and the public." He added that "crime and disorder and fear are important...things to measure, but there is a range...of other things that tell us something about what we do that makes a difference in our communities." Those things include "listening to the people we serve. Spending time [in neighborhood and community meetings] we hear things that are important to people," and these things may not necessarily fall under categories such as fear and disorder. Some people, Stephens said, want to dismiss information that comes in through calls from citizens: "I think call information is of critical importance. It's a major untapped source of knowledge about what people care about in the community and where problems are actually at...." "Of greatest importance, if you really want to measure anything...you've got to focus real hard on neighborhoods. You've got to build from the neighborhood up.... If we can focus our energies and efforts in policing nationally on doing the best we can with neighborhoods and working on problems with those neighborhoods, then we can, in fact, understand a lot more about what we do and the impact of what we do and then build up from there to a citywide perspective." Taylor's presentation was a summary of his paper, "Going over Grime Reduction: Community Policing Initiatives and Signs of Incivility." The paper provides an example of the considerations, details, and terminology involved in the researcher's thorough attempt to address something that matters. Taylor's subject is community disorder, which he calls "a broad and at times elusive concept, referring to social and physical conditions and events in a community, beyond the serious [Part I] crimes occurring there." His goal is to "address different conceptual, measurement, and policy-related issues surrounding various measures of community disorder." At the heart of his paper is Taylor's consideration of four different ways of assessing signs of incivility: survey based, onsite assessments, archival, and ethnographic. He describes the practical advantages and disadvantages of each. His conclusion is that these measures generally score well on reliability benchmarks (internal consistency and inter-rater reliability among raters), but, in the language of social science, "important questions related to construct and criterion validity remain. Convergent validity, discriminant validity, and criterion validity have not been established." Taylor identifies several questions "needing answers.... When people are telling us about perceived problems in their community or on their block, what do these reports represent? When onsite raters assess physical and social conditions on urban blocks, are those conditions so closely tied to neighborhood structure that they should be viewed as just reflecting structure? Or are they capturing something different?" The discussion that followed the presentations amounted to a 3-hour bouillabaisse, in the words of Francis Hartmann, the meeting moderator and executive director of the Program in Criminal Justice Policy and Management at Harvard University's Kennedy School of Government. The ingredients included many perspectives on what matters. Following are excerpts from the discussion. About Communities Michael E. Clark, president of the Citizens Committee for New York City, said that, "From the community perspective...on the subject of what matters, I see at least four clear things that jump out.... First [are] serious crimes that also cause community disorder.... Second are less-than-serious crimes that also cause community disorder.... Third is fear of crime.... Last...is noncriminal disorder.... I fear like a lot of people that too much of what we now call community policing has come to be focused on number four [cleaning up parks and other public places, graffiti removal, etc.]...as opposed to what the community is really worried about." Clark said, "The biggest thing that strikes me is the failure to apply some fairly simple, very well-developed market research tools" to measure what worries the community. "If you had an ongoing panel of citizens to talk to you about performance and victimization and other issues, including trust, I don't see how that's so enormously expensive or beyond our powers. I think it would be enormously powerful to start to move us toward what matters to the community...it would allow communities...to begin to assess the performance of their [police] precincts. If you think of policing as community service [there ought to be measures to assess that service]...and you ought to put those measures in the hands of citizens who pay the bills." Warren Friedman of the Chicago Alliance for Neighborhood Safety said, "Communities seem to me very involved, absolutely critical to making safer neighborhoods." But they were not reflected in papers prepared for the meeting "as something to measure, something to look at, something as critical." David Duffee, professor of criminal justice at the University at Albany, said he was bothered "by two large absences" in papers prepared for the meeting. He said there was "a lot of discussion about crime, disorder, and fear in relationship to some rather immediate structural characteristics of the community such as class and racial composition.... But I don't see any discussion about those larger forces" in struggling communities "which have a lot to do with where capital is located, what large economic elites decide to do with cities in terms of dislocating groups of people, housing policies.... We need to get some discussion about those very large forces that create a great deal of disorder...in order to ask whether the police can have any...lasting impact on disorder in the community." Duffee also said, "There ought to be some discussion about the traditional role of the police in the community [that] historically [was] to preserve the order that we have even though that order includes the fact that there are going to be some very disorderly neighborhoods. The police have often been asked to replicate or maintain the economic and political order as it stands...." Robert Sampson, professor of sociology at the University of Chicago, discussed community policing. He said, "The real issue for community policing, at least one major issue, is the extent to which not so much are the police able to pull it off but how are they able to bring the community into it in the sense of helping or leading the community to take responsibility itself for many [problems]." Clarence Harmon, former chief of police, St. Louis, Missouri, questioned "the utility of thinking only of the community in terms of citizens and the police. I think there is a larger dynamic made up of governmental and public institutions that impact on what the police do.... For instance, what can happen when you try to measure the impact of community policing in St. Louis in some neighborhoods...is that there is lower citizen satisfaction with the process merely because the jails simply can't confine [offenders and they are back on the streets]. My point in a broad, sweeping way is that you've got to consider all the components that make up the...entities that impact on the ability of the police to be successful at whatever outcomes we define for success." Johnnie Johnson, chief, Birmingham, Alabama, Police Department, said, "Nobody has any problem with me going through my processes as long as they produce the community's expected outcomes. If they don't, then I need to review my processes. I told the mayor, `Mr. Mayor, I want to look at whether or not I do what I said I am going to do.' He says, `Johnny, I want to know how many fewer homicides you're going to have, how many fewer robberies you're going to have.... That's what our community needs.' When it comes down to working at home, we have to deal with our communities. And inside the city, there [are several] communities. I've got one community that's concerned about burglaries...and another that's concerned about speeders.... The other question that comes up is, `How has the onset of drugs had an impact on solvability?` Those are a lot of issues that've got to be measured, but, at the same time, there's got to be a practical approach to policing." Robert Ford, chief, Port Orange, Florida, Police Department, said, "The diversity within communities is phenomenal. Our belief now is that every time we go to solve a problem, we create yet another problem for ourselves from another group. So, for example, if the senior citizens are concerned about youth groups gathering, we solve that problem, and that creates another problem with parents of youths who feel we are picking on the kids.... When you get to the more mid-level problems, the extent of the diversity becomes clearer." Peter K. Manning, professor of criminal justice at Michigan State University, raised another issue about measuring the reactions of communities. "One of the difficulties involved in the measurements that have been done...is that they are aggregated measures which do not take into account political, cultural, social realities.... Unless and until measurement is based on...some sense of community boundaries -- whether it's political or moral boundaries within that group of people who live there and interact -- it doesn't make any sense because you're collapsing apples and oranges, horses, mules, and so on. You're not measuring the same thing. So that any study has to begin with some kind of cultural articulation of what the meaningful boundaries of people's neighborhoods are. Until that's...sensibly done, and built into our understanding and analysis, we will continue to have this aggregated data which are about as useful as some...census data." Carl B. Klockars, professor of criminal justice at the University of Delaware, considered communities and order maintenance. "I've heard discussion about how we get the community involved.... There is another way to ask that question...namely, the community asking in what do we want to get the police involved.... The proper locus [for] responsibility for disorder in anything which resembles a real community...is the community itself. That's the place it's got to happen. I think it may be misguided on the part of police to claim responsibility for that [order maintenance] within communities.... Communities have to understand that they're the ones who can do something about" disorder. Mark H. Moore, professor at the Kennedy School of Government, raised a separate question about measuring communities: "Do we aggregate results inside a police department at the level of individual communities or citywide? ... It would be helpful to partnerships at local community levels with police departments if we somehow or other had the capacity...to aggregate at microcommunity levels as well as citywide.... To the extent that we decide that's an important thing to be able to do, that will increase the burden on reporting and measuring systems because it's one thing to capture performance at a citywide level. It's even more expensive and difficult to capture performance at a more microlevel, but it may be terribly important that we do that and we just suck it up and pay the money to get the job done." Bennie Click, chief, Dallas Police Department, said he was "not so sure that aggregate evaluation does a lot.... Once officers...identify [a] neighborhood [and] that neighborhood identifies with those police officers, it becomes much easier to measure within that small, that very small geographic area, the impact of changing something, doing something within that particular neighborhood to...reduce crime or whatever it is.... On a citywide level, I am not sure what aggregate numbers tell you." Mark Moore also observed, "One of the ways in which we have brought the community into the discussion of what matters is by -- almost without noticing it -- enlarging our concept of what matters from crime to [include] disorder and fear.... Now that's a very significant change.... Measures of [police department] performance...include [not only] crime reduction and victimization reduction [but also] reduction in fear and reduction in disorder." Dennis P. Rosenbaum, professor of criminal justice at the University of Illinois at Chicago, discussed measuring not only community reaction to police performance but also the reaction of individuals. An NIJ-funded project he is directing is "trying to measure police performance at different levels. Historically, we've been fairly good at asking general questions in the community about...how satisfied they are with the police overall, what kind of job they are doing, but we're also now beginning to look more at individual contact between citizens and officers and how they felt about that contact, whether it was fair, or how it was handled." About the Process of Policing Stephen D. Mastrofski, professor of criminal justice at Michigan State University, said that "both to police officers and the public, what matters a whole lot is the process of policing as well as the outcome.... When the crime rate goes up, police chiefs don't usually lose their jobs. When fear goes up, they don't usually lose their jobs. But they can lose their jobs when there is a riot [or]...a corruption scandal. When you ask the average citizen why he liked or didn't like some particular aspect of policing or how do you feel about the police, time and again they tell you it's how they were treated. I think that at some point [in] what matters we have to [include] the kinds of things the officer on the street and the citizen can pretty well assess: Was I treated well or wasn't I? Was I treated with civility, respect? ... [There's a] whole list of things about the process that matter tremendously...and is largely absent from this discussion." Roger B. Parks, professor in the School of Public and Environmental Affairs at Indiana University, also discussed the notion of process, fairness, and lawfulness. "How people are treated -- these are really measures that matter.... Police chiefs...do lose their jobs because of individuals who were involved in processes with citizens that turn out very unfortunately.... Police chiefs spend a tremendous amount of time being concerned about this and training officers to behave in ways that are lawful, that are fair, and that are appropriate in encountering citizens.... It's important to have some measures of that as what matters because this is crucial to continuing leadership." Chief Robert Ford of Port Orange said his department has examined "what was successful in increasing citizen satisfaction.... One of things we found [is that] citizen satisfaction is extremely highly correlated with police process, with the officers spending at least 10 or 15 minutes...[going] out and [trying] to do something. And the citizen always came to the same conclusion: `I don't care whether you succeeded, but the guy tries.'" Stuart Scheingold, professor of political science at the University of Washington, commented on how combating disorder may affect police process. "When you start dealing with disorder, you are intruding and intervening in people's lives short of any law violation. It seems to me that at some point it is at least possible...that the police are engaging [in] those kinds of behavior that are likely to create the kind of process concerns...that we have talked about.... Is this a realistic kind of problem? Is there...a tension between some crime attack-disorder attack strategies and process concerns about fairness and lawfulness?" Sally T. Hillsman, Deputy Director of NIJ, picked up on the subject of measuring police process. A serious question for the group to discuss, she said, was "whether or not something such as public trust and confidence is indeed an outcome that we want to measure directly or whether it's just a kind of offhand measure of process.... Is access to police service a process issue that we should relegate to stuff that we do when we have simply counted crime and measured...disorder? Or are these fundamental outcomes that police in their professional goals really wish to achieve? I think that is a fundamental discussion and debate. We could have a list of about six different things.... Is integrity an outcome? Is fairness an outcome? ... If those things are outcomes, then they are worthy of central measurement questions." For Mark Moore of the Kennedy School, "It's very important that many of the things we treat as process measures be accorded the status and significance of outcome measures." He makes these distinctions "within a broad category of process measures." 1. "Service quality at the level of individual contacts with citizens." 2. "Responsiveness of the police department to aggregated community groups." 3. "Lawfulness.... It might not be that there is any particular constituency [for it]...and yet it would be something that...most of us would be loath to give up and would like to find some way to measure." 4. Related to lawfulness, "a kind of fairness that has to do with distributing resources and protection equally and fairly across the community...an aggregate characteristic of a police department that might be important to measure as well." 5. "The standing of the police department in the community's eyes as an agency that's available...to all, that's perceived as operating fairly and effectively." About Expectations About the Police Charles P. Austin, chief, Columbia, South Carolina, Police Department, discussed people's expectations of the police. "When we evaluate the service delivery process and what we expect the outcomes to be, there are several issues that we take a look at from the practitioner's standpoint. They include cultural issues, geographic, social, economic, and political considerations.... Typically, the people who place the greatest demands on our resources and our services do so in situational circumstances so you can never really get an accurate grasp as to what is really expected. We will hear one set of concerns in this room, then go back to our respective jurisdictions and go into the various communities and you will hear different sets of concerns depending on what has happened most recently. Therein lies the greatest difficulty that we find ourselves confronting." Chief Darrel Stephens of St. Petersburg commented, "In our case, we always have to work with the neighborhoods in terms of establishing priorities, and that process...is instructive for both the police and the people in the communities because...people in the community [have expectations of the police] that aren't very real, and a lot of ideas we have about people in the community...aren't very real either. That process of...identifying problems and establishing priorities is...important to what outcomes actually turn out to be...." Jack Greene, professor of criminal justice, Temple University, noted a "large concern in the public sector about the quality of service received by two groups.... There is a user base. Did you get the service? Was it good? Did it solve your problem? There are routines that many agencies which are quality focused on service actually deliver... and one can measure that in straightforward ways. And then there are the symbolic components of government. Whether it makes people feel good, makes them not feel good; makes them feel safe, makes them not feel safe...." In Los Angeles and Philadelphia, two cities where Greene has done a lot of work, he found that the concerns of citizens are "either the police don't show up at all or show up considerably late or very late, and then when they get there, they're uncivil.... [And in a lot of cities] there is a group of populations [largely minority] that feels that it is underpoliced...[and doesn't] get services to which [it is] entitled. When that service does dribble down...it's hostile. So equity issues...are very much outcomes in this process." Greene suggested ways to measure these issues. "It seems to me that one can focus on incidents, problems, situations which are aggregate measures...disaggregated to the community level over time. One could look at user-based calls to cops, and what do they get once they call them, and the larger civic base which is what is their support for city government, and how do the police contribute to your sense of dependence. All of those measures are reasonably defined. It's a question of getting them all together in the same place, and then there's the huge logistic of managing that data collection." Jean Johnson, director of programs, Public Agenda, said that most people, "unless there's a really dramatic increase or decrease in crime, [will] not be measuring effectiveness by your evaluations, however you design them. They will be measuring them by individual encounters with police.... Do the police catch the bad person who did something terrible, and are they successful in getting them out of the neighborhood?" Andrew Benson, reporter, Cleveland Plain Dealer, noted that "a lot of people have talked here about community meetings [and] one-on-one interactions [with the police]. But a lot of people don't have any contact with police departments.... They get their perceptions from somewhere else...through the mass media.... That gets us back to...the old measures that are very simple, very alluring...which is the UCR: Crime is up or crime is down compared to last year, compared to last month. And that's something that journalists can get a hold of and run it. That's something politicians can get a hold of...." Should We Expect Police Activities to Impact on Measures of Crime, Disorder, and Fear, and How Will We Know? The initial response to the second question in the day's discussion came from then-Commissioner William Bratton of New York City, where major felonies are down 26 percent since he took over the city's police department 2 years ago. Bratton and the department have claimed an aggressive share of the credit for the drop in crime to levels not seen since the early 1970s. He first addressed the question by turning it around. "If we don't expect police activities and police departments to have an impact on crime, disorder, and fear, they almost certainly won't," Bratton wrote in a paper prepared for the meeting. "By accepting the prevailing image of police departments as slow moving and relatively ineffectual bureaucracies and by assuming that nothing can be done to change them, we are, in effect, making a self-fulfilling prophecy. No organization...is going to achieve high-performance results in an atmosphere of such low expectations." The paper says that alternative explanations such as demographic or economic factors or incarceration rates cannot account for the drop in the city's crime rates. The high-performance results he champions have meant that crime is down citywide in all 76 precincts between 15 and 40 percent, Bratton said. In the discussion the commissioner said that "what we're engaged in, in New York City, is results-oriented management. The results I'm looking for in New York are, quite simply, reduced crime, reduced quality-of-life negatives such as disorderly behavior, and, with that, reduced fear. We reorganized the New York Police Department to focus on producing those results. Not the 1, 2, 3 percent results that we were seeing in crime decline in the [early] '90s in New York City but now the 26 percent in our reduction in crime over the last 2 years." He added, "We [the police] can control crime...within limits and under certain conditions. In New York City, we were able to do it for a variety of reasons. One, politically...the mayor...is very supportive of the issue. Two, the number of police [38,000] that I have.... Other factors -- community involvement, community policing. New York is incredibly a community-involved city.... It is made up of hundreds of neighborhoods, and they are involved politically and involved with police precincts. They are in the game." Bratton also addressed researchers and academics: "The writings of the last several years have raised a concern on my part that the police were being discounted as an effective force in our ability to not only reduce crime where it was high...but to control it, to keep it at manageable levels.... Increasingly you are completely discounting the ability of the police and others -- with the information you are providing -- to make a difference. In that respect, you're scaring the hell out of people. There is a sense in this country now that no matter what we do over the last 5 years of this decade, [we] are going to [have] a blood bath.... You are doing a disservice to the public...." Two other background papers were prepared for the discussion of the second question, by Alfred Blumstein, a professor in the H. John Heinz III School of Public Policy and Management at Carnegie-Mellon University, and George L. Kelling, professor of criminal justice at Northeastern University. Of Bratton's presentation, Blumstein said, "The whole theme of the issue that I think Bill Bratton has caught everyone's attention with is the basic theme that's sweeping industry of continuous improvement. One can do [continuous improvement] in a variety of ways. One can hold one's managers' feet to the fire based on the outcome measure you've defined, and the outcome measure certainly should include...various kinds of crime measures.... That works in that jurisdiction where excellent management is in place. One of the reasons we are all together in Washington under NIJ's sponsorship is that we really want to develop a knowledge base and an understanding of the mechanism that works since what happens in New York can be applicable to other places beyond a `Do as I do' kind of repetition." "We're clearly seeing something in the big cities that represents a decline in some of the most serious crime, and the smaller cities are lagging, in part, I believe as speculation, because of the lag in the arrival of the drug markets, which I think were a major factor in the growth of crime in the late `80s. Drug markets arrived in the smaller cities 1 to 5 years later. This phenomenon [the decline of big-city crime rates] may be a maturation of drug markets, and the maturation will occur later in the smaller cities." Kelling saw merit in Bratton's explanation for the decrease in New York's crime rate. "I believe something very powerful is happening in New York City and in some other cities. My own interpretation...might be the same as Bill's.... [In New York City over the last 20 years] we've seen the development of a powerful community movement...really an alternative paradigm to the criminal justice system paradigm and the law enforcement paradigm.... There's a movement afoot to move away from the idea that crime control is the responsibility of the professionals." Then, Kelling said, along came a manager, Bratton, "who suddenly said to the city, `By the way, on top of your activities...on the top of your concerns...private security...community crime control...I am going to energize 38,000 cops and get them to be pretty active. Under those conditions I can understand that the police are having a powerful impact, and that's my interpretation of what's going on." Following are excerpts from discussions at the meeting that focused on Bratton's claim. Ronald Clarke, dean of the School of Criminal Justice at Rutgers University, said he was "actually very glad that Mr. Bratton has taken on responsibility for reducing crime rates. I think that the problem with the control and prevention of crime over the last 15 or 20 years is that nobody has been willing to take on the responsibility for reducing crime. There are all kinds of constituencies that, according to other people, have the power or the capacity to control crime, or there are people saying nothing can be done about it. I think that has lead to a lot of very fuzzy thinking about causes of crime.... If the police stand up and say, `We can control crime,' I think that will provide the leadership that the community, that society, wants." But, Clarke said, "I personally doubt...I can't produce any real evidence...that police action can have produced such an enormous drop in crime" as Bratton takes credit for. "Police action can impact crime in small localized settings if there is a clear...focus," he added. Dennis E. Nowicki, chief, Charlotte-Mecklenburg, North Carolina, Police Department, said of the causes of the crime decrease in New York, "All the things that we [not just the police but the community, other agencies of government, the business community, and neighborhoods] do are at play in crime reduction.... Bill's looking at much more recent information to try to figure out what's going on, and maybe he needs to look back 10 or 15 years ago to see what went on then" to determine the behavior of children who now are adults in the crime-committing age group. "Most of my [crime] prevention initiatives are targeting youngsters...and the risk to me is that that's not going to pay a dividend while I'm around. It's going to pay a dividend maybe 10 or 15 years down the line when a Bill Bratton-type comes into Charlotte and takes credit." (Laughter ensued.) Nowicki said also that statements that "the police can't have an effect on crime don't compute with me.... We overstated for many years the lack of effect that we have on crime. What you're being challenged on now is that overstatement. We do have an effect. But we're not the only ones that have an effect." Chief Johnnie Johnson of Birmingham discussed the impact of moral education on crime rates. "I want to believe that if the police are operating at 100 percent, we can affect crime.... However, I'm very, very skeptical about accepting the responsibility of that impact because what happens when it [crime] goes up. Do I still say, `Well, I must have faltered somewhere or my folks have stopped working. They sit down on me and that's why it has gone up.' If I work my people at 100 percent, will that stop [someone's] son from getting out [and] burglarizing somebody? Will that stop this family that I know for the last 10 years has produced criminals to stop producing criminals.... That's a thing we have to deal with.... "I do have to give credit, though, to Miss Smith who tells Johnny, `Johnny, you will not steal. That's wrong.' I can't take that credit from her because she's having an impact on the crime out there. I have to give credit to that minister...that cautions his congregation about getting involved in crime and use of drugs.... It's a total community effort that has the most impact on the occurrence of crime." Warren Friedman of the Chicago Alliance for Neighborhood Safety said, "I think the police can have an effect [on crime], but I have some worries about that. The kinds of worries come from experience with community organizations who...have shut down drug houses, opened up parks, and can do that with only sporadic support from the police and usually support which has to do with confrontation which gets their attention. That's the history of Chicago up to a short time ago.... I suspect that now that there's beginnings of partnership between police and the community in Chicago, that's a much more powerful tool to shut down targeted areas, open up parks.... But I also know from the community experiences that it's very difficult to sustain. You can win the battle the first time; you can point at a place where there's no activity, and you come back 6 months later and it's there. It's there because the community couldn't stay organized and sustain their effort and because the police are very busy and are moving all over the place." Responding to Bratton's remarks, Chief Robert Ford of the Port Orange Police Department said, "If you study the variance between police departments in crime rates...and if...you group police departments by census variables, [and] if you take all the variables [number of police per capita, demographics, number of people living in a place over 5 years, numbers of juveniles per thousand population]...and put them all together, and you throw them into a beautiful matrix, and you factor them, and you do some beautiful multiple correlations, what you have left is still a huge amount of variance in the crime rates.... I'm saying that better departments produce better results." Edward Flynn, chief, Chelsea, Massachusetts, Police Department, said, "Somewhere along the line you guys [researchers and academics] went off track and we had to keep plunging along in this morass of crime, disorder, and community deterioration.... What Bill [Bratton's] people have been doing, which...those academics who do some of this criminological research could do a wonderful job of, Bill's people debrief arrested suspects. And they find things about how the crime occurred. Now I know factors cause crime, but people commit it. And if you want to hold police accountable for diminutions in the crime rate, then I challenge you to help us find out how crimes are committed...." Tom Koby, chief of the Boulder, Colorado, Police Department, said, "I think there is a basic misunderstanding here about what Bill said.... [He said] police -- not law enforcement but police in the sense of the grand tradition of police -- in partnership with an empowered community can impact crime and disorder." Aric Press, a senior writer for Newsweek, noted that some police officials and agencies now were taking credit for decreases in crime where, recently, that was not the practice. Press said, "The rhetoric, it seems to me, began to change a few years ago where we were all told that crime was too big just for the police. The community had to be involved, that there were all sorts of factors involved in it. It was beyond the power of the police to affect directly [crime rates] too much. Now we have this remarkable success [declining crime rates in New York City and some other cities]. The one thing I say to you...is that I hope that success continues because, if not...if that's the core business [of policing] -- how many homicides, how many robberies, how many burglaries up or down -- what's been said at the table a few times, about how people don't lose their jobs because of UCR [Uniform Crime Report] measures, that's going to change. If that's the way you want to be measured, that's a tiger you're riding." Robert H. Langworthy, professor of criminal justice at the University of Cincinnati, on academic leave to work with NIJ during 1995-96, said of drops in crime rates in New York and other cities: "The dilemma that we are in right now is that we have an empirical fact in search of an explanation.... It is an empirical fact that in some cities crime has gone down. We have a lot of explanations that we all have been led to at one point or another that don't seem to allow us to account for that decline.... One of the problems that we have with agencies trying to claim success is that they don't look across the aggregate. We don't know what...police practices have...experienced a decline in crime. We don't know what similar police practices that are being employed in New York City have been associated with increases in crime. But we do know...that in at least some places crime has gone down. We also know that it has gone up in other places.... And we are thrashing around for an explanation.... What we have is an empirical fact and we are searching for an explanation. I think it is far too premature to presume." Jack Greene of Temple University said, "Part of the explanation of New York's experience is...that crime fell off the plate for about 10 years. We were all fixing broken windows and broken houses. We weren't fixing the crime problem that was out there. And forget all the analytical and theoretical linkages there. Crime got back on the agenda. And most police chiefs today use community policing to talk about crime issues where they didn't [5 or more years ago]. I think that's an important shift.... What Bill Bratton's paper does is get the police back in the lead.... Somebody has got to be accountable and responsible for at least keeping this in focus." Moderator Francis Hartmann asked about developing "indices that might help the chiefs and...the researchers." Jack Greene replied, "I don't think we can lose the UCR or the victimization surveys or other aggregate indices of how we are doing in a particular community. They are baseline information, they have been collected over a long period of time, and they are more stable than people give them credit for." But he added that "there are levels of analysis and data sets that already exist at several levels that we do not exploit very well. Calls for service data [for example]...do not get analyzed very well in a lot of places.... [h]undreds of thousand of calls for service...create all kinds of problems and all kinds of demands for police resources. [So] aggregate data exist that need to be analyzed in a different way. Further, there is [a] moral, ethical kind of value-based assessment that can go on in an annual or biennial survey of the citizenry. There are city police departments that print [surveys] on the back of water bills...a `how-are-we-doing' questionnaire. That goes, at least, to everybody who owns a building.... So there are ways to collect information in fairly unobtrusive ways." Robert H. Langworthy cited the availability of Law Enforcement Management and Administrative Statistics (LEMAS) that are gathered every 3 years under the auspices of the Bureau of Justice Statistics. Data on personnel, expenditures and pay, operations, equipment, computer and information systems, and policies and programs are compiled from State and local agencies with 100 or more officers. "I am a devoted fan of the LEMAS surveys," Langworthy said. "LEMAS is an incredibly valuable vehicle for providing information about policing." Of course, he added, "Any survey of that sort is going to have a huge number of problems.... But it will get better with age as do all these enterprises. The problem, though, is that we still don't build into the process when we design the LEMAS system...what it is the police are supposed to be doing that's going to have an impact.... It's a fairly simple enterprise, I suspect, but it needs some attention." ------------------------------ Selected NIJ Media Products Listed below are some recent National Institute of Justice media products of interest to readers of "Measuring What Matters." These products are free, except as indicated, and can be obtained from the National Criminal Justice Reference Service: telephone 800-851-3420, e-mail askncjrs@ncjrs.org, or write NCJRS, Box 6000, Rockville, MD 20849-6000. The documents also can be downloaded through the NCJRS Bulletin Board System or at the NCJRS Anonymous FTP site in ASCII or graphic formats. They can be viewed online at the Justice Information Center World Wide Web site. Call NCJRS for more information. Please note that when free publications are out of stock, they are available as photocopies or through interlibrary loan. Blumstein, Alfred, Ph.D., Youth Violence, Guns, and Illicit Drug Markets, VHS videotape, 1995, NCJ 152235, U.S. $19, Canada and other foreign countries $24. Mastrofski, Stephen D., Research Preview, Law Enforcement in a Time of Community Policing, July 1996, FS 000149. Roth, Jeffrey A., and Mark H. Moore, Reducing Violent Crimes and Intentional Injuries, Research in Action, October 1995, NCJ 156089. Sampson, Robert, Ph.D., Communities and Crime: A Study in Chicago, VHS videotape, 1995, NCJ 156924, U.S. $19, Canada and other foreign countries $24. Sherman, Lawrence W., Ph.D., Reducing Gun Violence: Community Policing Against Gun Crime, VHS videotape, 1995, NCJ 153730, U.S. $19, Canada and other foreign countries $24. Skogan, Wesley, Ph.D., Community Policing in Chicago: Fact or Fiction?, VHS videotape, 1995, NCJ 153273, U.S. $19, Canada and other foreign countries $24. A summary of the presentation is available free; ask for FS 000105. Taylor, Ralph B., and Adele V. Harrell, Physical Environment and Crime, Research Report, May 1996, NCJ 157311. ------------------------------ What Indicators Measure Police Performance? by Jeremy Travis, Director The police play a central role in our democratic society. They are called upon to enforce our laws, to observe constitutional restraint upon the exercise of governmental power, to answer individual calls for help, and to respond to community demands for safety. Of all governmental operations, the police function is the most intimate -- the daily, varied encounters between police officers and individuals, ranging from routine to traumatic, represent the most visible and powerful interactions between the government and the public. If the police perform their role effectively, our society benefits immeasurably; if the police perform their role poorly, the damage to public confidence and democratic principles can be irreparable. How do we know whether the police are doing well? To encourage discussion, debate, and solid research on this question, the National Institute of Justice, in collaboration with the Department of Justice Office of Community Oriented Policing Services, convened a Policing Research Institute discussion entitled "Measuring What Matters." We decided to bring together, in three 1-day sessions, about 40 police executives, leading researchers, community leaders, journalists, and government officials to discuss the challenges of assessing police performance. This report presents highlights of the first session. In this session, we sought to capture the national debate that has accompanied the welcome news of sharp declines in violent crime rates in some large cities: Can the police claim credit for this change? What other factors might have contributed? How can we examine issues of causation in contemplating a phenomenon as complex as criminal behavior? We confronted another critical issue: What other indicators should gauge police performance? Fear? Disorder? Public satisfaction? Public trust and confidence? If these are additional indicators that the police are doing well, by what logic do we hypothesize that the police can affect these more ephemeral, but equally compelling, indices of societal well-being? These questions are being asked at a time of historic transformation in the way the police are thinking about themselves. The movement toward community policing, toward a problem-solving model of police services, requires a new set of understandings of the role of the police and, in turn, a new set of performance measures. Future sessions of the Policing Research Institute on "Measuring What Matters" will view these issues through different prisms. The second session will look at the police from the outside, examining public measures of satisfaction as interpreted by public opinion surveys, press accounts, and community perceptions. The third will look inward, examining internal measures of the performance of individual officers, units of officers, and entire departments. From all three sessions, we anticipate making recommendations on new ways of measuring police performance. It is our hope that these discussions, the widespread dissemination of the highlights, and a future compilation of commissioned papers on these topics will shed light on the critical police function at a critical time in our history. ------------------------------ The Decade's Most Important Criminological Insight "Probably the most important criminological insight of the decade has been the discovery in a very systematic fashion of repeat multiple victimization. This has tremendous implications both for criminological theory and...practice in the field. We knew already...of the concentration of crime in areas.... Take census tracts of the city and rank order them from low crime to high crime. That last 15 percent of census tracts accounts for 40 percent of your total.... You get this tremendous piling up of crime in places. It turns out that what appears to cause crime to pile up very heavily in high crime areas is to a certain extent because more people are victimized but, more so, because some people are victimized repeatedly. They may be commercial establishments, they may be organizations, they may be individuals, but that piling up of repeat multiple victimization is mostly what makes a high-crime neighborhood a high-crime neighborhood. This drives theory.... The question is why. ... Why in the same way that a small percentage of repeat offenders contribute a large percentage of crime [does] a small percentage of repeat victims contribute a large percentage of victimization. ... [it] also drives practice.... So this notion of finding ways of measuring on the one hand and, in policy terms, responding to repeat multiples...is one of the most important kinds of ideas out there that criminologists have to contribute to practice." -- Wesley G. Skogan, professor of political science at Northwestern University ------------------------------ The Perversity of the Measures We Use George L. Kelling, professor of criminal justice at Northeastern University, lamented "the perversity of many of the measures that we use in policing. It's not [only] that they're not just accurate but that many times they give exactly the wrong results." In his paper prepared for the meeting, Kelling cited his experience riding with police officers on bicycle patrol in a densely populated working-class area. "The trouble is that while some very profound things are happening in the wards in which I bicycled, arguably with community officers playing a pivotal role, officially -- that is, as represented in departmental records -- these officers are doing practically nothing. Moreover, given police traditions, the officers themselves are bereft of any language to discuss what they are doing. They still see themselves as responding to, or managing, incidents, and when probed about what they are doing or why they are doing it, their universal response is: `It was only common sense. ' Please do not misread me here. I am not suggesting that officers are stupid; I am suggesting that the rhetoric of policing, and the things police count, are still so tied to the reform, the so-called `professional,' model of policing that officers lack a context either to understand or to talk about what they do or what they accomplish. In the city that I am reporting on, the officers, the wards, and the police department suffer from the inability to count the problems with which officers deal or the outcomes officers achieve." ------------------------------ The Need for a Big-City Data Base "Something common is going on in the...big cities. We really want to know what [it is].... It will be very desirable...to get timely research results. That means a data base -- not just the dependent variable of crime rates -- but where they're changing, who committing crimes is changing, how those changes relate to the claim of aggressive stop-and-frisk interventions, the aggressive minor offense interventions. Inevitably those interventions don't happen uniformly across a city but happen selectively in different places. It would be very desirable to have those relationships articulated and explicated and sorted out. In doing this, we need both aggregation and disaggregation - - disaggregation by unit, locale, by subgroups committing the crime, by factors contributing to it. A major factor in the growth of violence has been youth violence. A major factor in that has been the presence of guns resulting from the drug markets and the diffusion among a lot of young kids.... We want aggregation by looking across the big cities and seeing what common trends [exist]." -- Alfred Blumstein, professor in the H. John Heinz III School of Pubic Policy and Management, Carnegie-Mellon University ------------------------------ Making Sense to Voters "We do need to know not only what matters but what can be said about what matters in a form that makes sense not only to policymakers but ultimately to voting constituents who can easily be swayed by manipulated misinformation." -- Edward Flynn, chief, Chelsea, Massachusetts, Police Department ------------------------------ Measuring What Matters, from City to City "Measuring what matters. What matters really does differ from city to city.... I know from my experience in Boston in the '70s, you lived and died based on your response time. ... In measuring what matters, it's...time relevant. In New York City at this time, what matters in the minds of people -- the press, the politicians, the public -- are crime rates.... Compounding the dilemma we're talking about here -- measuring what matters -- is really, every city in the country has different concerns at different times." -- William Bratton, former commissioner, New York City Police Department ------------------------------ Going Toe-to-Toe with Researchers "I come to this conference with a healthy skepticism, quite frankly, about what researchers and academics can provide us.... I don't see a lot.... Time has been wasted on some research that has been done.... We're now educated in the police field. We have people who can go toe-to-toe with the researchers in our organizations. It's not quite so easy to continue to bluff your way by us. I look forward to having some healthy discussion about the issue in front of us...." -- Tom Koby, chief of police, Boulder, Colorado ------------------------------ You're the Measuring Device "I'd like to challenge that. I think you measure every time you walk out there to talk to your dispatchers or your officers or your citizens. It's just that you're the measuring device.... You could take us a long way in this discussion if you would talk to us about what matters to you in those conversations. What are you listening for? What do you want to know from those people.... All of the key actors in the police community relationship need to come forward, sit down, and talk about what matters to them...." -- Mary Ann Wycoff, senior research associate at the Police Executive Research Forum, in response to Chief Tom Koby of the Boulder, Colorado, Police Department, who told the meeting he did not spend a lot of time measuring. ------------------------------ Matters of Measure: About Macro to Micro "Much of our difficulty in measurement stems from our effort to constantly use global, all-encompassing terms -- the trinity of crime, disorder, and fear -- which for me has no greater meaning than disease, illness, and the associated anxieties. We need desperately...to move from the macro to the micro in every possible way. I was pleased to see [in some papers prepared for the meeting] a substantial effort to move along those lines with references to discrete problems and to local measures. That's what the community is concerned about -- local measures. That's what the chiefs are concerned about...local problems, specifically defined problems where you can develop targeted responses and, as a result, increase substantially the potential for measuring the actual impact. One of the side benefits is that we move away from...trying to measure attitudes and perceptions to measure specific behavior and specific outcome." -- Herman Goldstein, law professor at the University of Wisconsin ------------------------------ Matters of Measure: About Racial Tensions In discussing measurement, Herman Goldstein said to keep in mind that, "Today...in policing, the bottom line is not crime, the bottom line is the racial tensions in our large urban areas, and there's no escaping that. We may, in fact, have an impact upon crime, disorder, and fear, but if, in the course of doing that, we substantially increase tensions in the community, we've made matters worse. It suggests to me that we need to move more aggressively in monitoring and evaluation to take the vital signs of the community...to reflect on what the impact is of different [police] strategies and practices that we adopt." ------------------------------ Matters of Measure: About Police-Crafted Measurements Herman Goldstein cited the measurement efforts of police practitioners "in dealing with problems on the street. They learn from the bottom up. For example, officers in San Diego who confronted the classic long-term problem of street prostitution came up with some incredibly innovative measures of their impact, of how much business the hotels were doing, how many hours stores were keeping open, how much they were investing in security. These [measures] admittedly are amateurish, they're baby steps, and they tend to invite dismissal and disdain...because they don't meet all of the standards. But they are very significant. These officers are uninhibited by all kinds of concerns that we have.... They ask the right questions, and we have a lot to learn from them. I think there is a way to capture that common sense...." ------------------------------ Indicators That Say "Try Again" "We can't spend the money that is needed for detailed, deep evaluations on every innovative idea people may have. But we need to have indicators at some minor level that say, `Try again, try again. Good idea. Look at it some more.' We always will be looking for simple, aggregate measures and then look at some things in detail and be able to move between them. I hope this session and the ones that follow will begin to give us a way to find links so that academic researchers can start to see how can we take these measures and find proxies that any community can afford to produce. And practitioners can begin to see that it is really important to understand what measures are developed and not be the passive recipient." -- Patricia Brantingham, professor of criminology, Simon Fraser University ------------------------------ Papers prepared for this and the other Policing Research Institute meetings are scheduled to be published in book-length form by NIJ after the third session. ------------------------------ Data Sources Following are data sources mentioned in this report. Uniform Crime Reports (UCR): An annual statistical summary, collected and published by the Federal Bureau of Investigation, of crimes reported to the police. The UCR contains detailed information on Part I, or serious, offenses: criminal homicide, forcible rape, robbery, aggravated assault, burglary, larceny, car theft, and arson. National Crime Victimization Survey (NCVS): Surveys a large sample of U.S. residents to determine annual victimization rates of all Part I crimes except homicide and arson. The NCVS sample, prepared for the Bureau of Justice Statistics, consists of about 50,000 households and 101,000 persons who are interviewed every 6 months for 5 years. Law Enforcement Management and Administrative Statistics (LEMAS): Data collected from State and local agencies with 100 or more officers on personnel matters, expenditures and pay, operations, equipment, computer and information systems, and policies and programs. LEMAS is compiled about every 3 years under the auspices of the Bureau of Justice Statistics. ------------------------------ Participants in the Policing Research Institute Meeting, November 28, 1995, Washington D.C. Geoffrey P. Alpert Professor of Criminology School of Criminal Justice University of South Carolina Charles P. Austin Chief of Police Columbia, South Carolina, Police Department Andrew Benson Reporter Cleveland Plain Dealer Alfred Blumstein J. Erik Jonsson University Professor H. John Heinz III School of Public Policy and Management Carnegie-Mellon University Joseph E. Brann Director Community Oriented Policing Services U.S. Department of Justice Patricia Brantingham Professor School of Criminology Simon Fraser University William Bratton Police Commissioner New York City Police Department Michael E. Clark President Citizens Committee for New York City Ronald Clarke Dean School of Criminal Justice Rutgers University Bennie Click Chief of Police Dallas, Texas, Police Department David Duffee Dean School of Criminal Justice State University of New York at Albany Michael J. Farrell Deputy Commissioner Policy and Planning New York City Police Department Edward Flynn Chief of Police Chelsea, Massachusetts, Police Department Robert Ford Chief of Police Port Orange, Florida, Police Department Warren Friedman Chicago Alliance for Neighborhood Safety Herman Goldstein Professor School of Law University of Wisconsin Jack Greene Professor Department of Criminal Justice Temple University Clarence Harmon Chief of Police St. Louis, Missouri, Police Department Francis X. Hartmann Executive Director Programs in Criminal Justice Policy and Management Kennedy School of Government Harvard University Sally T. Hillsman Deputy Director Office of Research and Evaluation National Institute of Justice U.S. Department of Justice Jean Johnson Director of Programs Public Agenda Johnnie Johnson Chief of Police Birmingham, Alabama, Police Department George L. Kelling Professor College of Criminal Justice Northeastern University Carl B. Klockars Professor Division of Criminal Justice University of Delaware Tom Koby Chief of Police Boulder, Colorado, Police Department Robert H. Langworthy Professor of Criminal Justice University of Cincinnati Program Manager (former) Office of Research and Evaluation National Institute of Justice Peter K. Manning Professor School of Criminal Justice Michigan State University Stephen D. Mastrofski Associate Professor Administration of Justice Pennsylvania State University Phyllis McDonald Social Science Analyst National Institute of Justice U.S. Department of Justice Mark H. Moore Professor Kennedy School of Government Harvard University Dennis E. Nowicki Chief of Police Law Enforcement Center Charlotte-Mecklenburg, North Carolina, Police Department Roger B. Parks Professor School of Public and Environmental Affairs Indiana University Aric Press Senior Writer Newsweek Dennis P. Rosenbaum Professor Department of Criminal Justice University of Illinois at Chicago Robert Sampson Professor Department of Sociology University of Chicago Stuart Scheingold Professor Department of Political Science University of Washington Wesley G. Skogan Professor Department of Political Science Center for Urban Affairs Northwestern University Darrel W. Stephens Chief of Police St. Petersburg, Florida, Police Department Ralph Taylor Professor Criminal Justice Department Temple University Jeremy Travis Director National Institute of Justice Benjamin B. Tucker Deputy Director for Operations Community Oriented Policing Services U.S. Department of Justice David Williams Assistant Chief Portland, Oregon, Police Department Gerald Williams Director Bill Blackwood Law Enforcement Management Institute Sam Houston State University Jane Wiseman Assistant to the Director for Strategic Planning National Institute of Justice U.S. Department of Justice Mary Ann Wycoff Senior Research Associate Police Executive Research Forum ------------------------------ Discussion Themes Several themes emerged in the discussions that are covered in this report. They had to do with communities, the reduction in big-city crime rates, the process of policing, and expectations about the police. Communities. At least four themes about communities were heard: o The importance of measuring neighborhoods, the small geographic units that make up cities, to understand the impact of police endeavors. "You've got to build from the neighborhood up," said one participant. o The political, cultural, and social realities of communities and the larger economic and other forces that form them. These matters are often not considered or measured in evaluating the police in their interaction with communities. o The need of communities to be involved in resolving their problems, notably disorder. Communities are "absolutely critical to making safer neighborhoods," said a participant. o The difficulties police have in addressing the needs within communities. The decline in big-city crime rates. A major topic in the day's discussions centered on explanations for the significant decline in crime rates in many large U.S. cities. The explanations included (1) aggressive, results-oriented management of police departments, (2) a maturation of big-city drug markets that means less violence, and (3) the coordinated efforts of communities, the police, and other agencies of government in crime prevention. The process of policing. How the police behave -- how they act toward citizens -- in carrying out their jobs was a recurring theme during the meeting. "Both to police officers and the public, what matters a whole lot is the process of policing as well as the outcomes," according to a participant. "Citizen satisfaction is extremely highly correlated with police process," with police officers attempting to resolve the concerns of citizens, said a police chief. As to measuring the police, "process measures [should] be accorded the status and significance of outcome measures," according to an academic researcher. Expectations about the police. What the public expects from the police was another frequently voiced theme. Public concerns depend "on what has happened most recently," a police chief said. A professor of criminal justice noted two aspects to citizens' expectations: Did users of police services get the service they expected, and did it solve their problem? Did users of police services feel more or less safe, better or worse, as a result of interaction with the police? A public opinion researcher suggested that most people measure the police by their individual encounters with them. A journalist said many people's expectations and opinions of police come from mass media reports of daily crimes. ------------------------------ Thomas V. Brady 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, Bureau of Justice Statistics, Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. NCJ 162205