First-Line Supervision and Strategic Decision Making Under Compstat and Community Policing

Abstract

Compstat and community policing are powerful movements in U.S. police reform. Although different in important respects, both are heralded for being strategic, or strengthening a police organization’s capacity to detect changes in crime and disorder and respond effectively. Using data from six focus groups, this study examines how first-line supervisors reported making decisions and offering guidance on crime and disorder problems. Its major finding is that co-implementation of these reforms had affected supervision unevenly. A challenge to policymakers and researchers is to reconsider how patrol supervision in co-implementing departments might be restructured to strengthen the strategic dimensions of both reforms.

Keywords

Compstat community policing first-line supervision police reform police decision making

Over the last quarter of a century, two of the most highly visible police reforms to have emerged in the United States are Compstat and community policing (Bratton & Knobler, 1998; Rosenbaum, 1994; Silverman, 1999; Silverman & O’Connell, 1999; Skogan, 2004, 2006a; Weisburd, Mastrofski, McNally, Greenspan, & Willis, 2003; Willis, Mastrofski, & Weisburd, 2007). Compstat is a highly focused strategic management system concentrating on reducing serious crime by decentralizing decision making to middle managers operating out of districts, by holding these managers accountable for performance, and by increasing the organization’s capacity to identify, understand, and monitor responses to serious crime problems (Eterno & Silverman, 2006; Henry, 2002; Kim & Maurborgne, 2003; Maple & Mitchell, 1999; Silverman, 2006). Community policing is characterized by a variety of justifications: strengthening public support for the police, building social capital, achieving a more equitable distribution of police services, and reducing crime, disorder, and fear of crime by customizing solutions to local problems (Rosenbaum, 1994; Skogan, 2008a). Its methods are similarly diffuse, but often include community partnerships, problem solving, and the delegation of greater decision-making authority to sergeants and their patrol officers at the beat level (Moore, 1992; Skogan, 2006a).

Both innovations have diffused rapidly throughout the United States and abroad (Skogan, 2006a, p. 5; Weisburd et al., 2003), and researchers are still trying to determine the effects of each of these reforms and their future prospects (Dabney, 2010; Weisburd & Braga, 2006). However, it is also important to know how these two reforms operate together when co-implemented in the same police agency. According to a 2006 national survey, 59% of large police agencies are pursuing both Compstat and community policing simultaneously suggesting how they work together has significant implications for how policing is done (Willis, Kochel, & Mastrofski, 2010).

This is the third article from a national assessment of this co-implementation issue. The first two used site-visit data to explain why Compstat and community policing operated largely independently and to make recommendations for their integration (Willis, Mastrofski, & Kochel 2010a, 2010b). The purpose of this article is narrower in scope. Its point of departure is the claim that both reforms are “strategic” or designed to heighten a police organization’s capacity to identify problems in its crime environment, detect any changes, and respond effectively. This strategic focus is then examined from the perspective of one rank, first-line supervisors, whose decisions play a key role in converting policy into practice (Skogan, 2008b).

Although there is a small amount of research on co-implementation, it is generally within the broader context of organizational change (see Willis, Mastrofski, & Kochel, 2010b). In comparison, there is virtually no in-depth research on the decision making of first-line supervisors in co-implementing departments. This is a significant oversight as presumably these reforms’ “big picture” approach toward handling an unstable environment should materialize in the kind of guidance that first-line supervisors receive from their superiors, in how they make decisions about crime and disorder problems, and in the kind of direction they give to the organization’s largest resource—its patrol officers. Thus, using focus-group data from six police departments, this article asks, “How, if at all, did the co-implementation of Compstat and community policing influence the strategic decision making of first-line supervisors?”¹

The answer to this question is important as studies often show significant slippage between the doctrines of particular reforms and how they operate in practice when implemented in a police organization (Clarke, 2004; Cordner & Biebel, 2005; Ratcliffe, 2002). Knowing how first-line supervisors said they adapted to the strategic components of these reforms can help identify where co-implementation problems are likely to occur and thus offer some direction on how they might be overcome. The next section describes how the key strategic elements of these reforms are supposed to work (at least according to their respective doctrines). This, then, acts as benchmark for making comparisons to how first-line supervisors said they made decisions within the context of these reforms.

First-Line Supervision in an Age of Reform

Before the emergence of community-oriented policing (COPS) in the early 1980s, patrol supervisors were described as performing traditional supervisory roles in organizations that were hierarchical, bureaucratic, and reactive (Engel, 2002, p. 52; Kelling & Moore, 1988). Sergeants were expected by their superiors to engage in preventive patrol, to ensure that their subordinates responded quickly and appropriately to individual incidents and calls for service, to check crime reports for inaccuracies, and to discipline officers who violated department rules (Allen, 1982; Manning, 1977; Rubinstein, 1973; Trojanowicz, 1980). This stress on internal control contributed to a supervisory system that was “essentially negative, relying primarily upon sanctions for non-compliance with police rules” (Weisburd & McElroy, 1988, p. 31).

Community policing and Compstat (which originated in the New York City Police Department in 1994) evolved in response to many of the bureaucratic pathologies associated with this traditional policing model (Bratton & Knobler, 1998; Reisig, 2010; Weisburd & Braga, 2006, pp. 11-12). Advocates for reform criticized police organizations for placing greater emphasis on “their organization and operation than on the substantive outcome of their work” (Goldstein, 1987, p. 236), for creating elaborate hierarchies and rules, for centralizing command and control, and for detaching themselves from the communities they served (Eck & Spelman, 1987; Trojanowicz & Bucqueroux, 1990).

To help overcome these challenges, reformers stressed the importance of refocusing the organization’s energies on reducing and preventing crime and responding to concerns identified by those outside the department. The reform literature on Compstat (Silverman & O’Connell, 1999) and community policing (Skogan, 2006a; Trojanowicz & Bucqueroux, 1990) suggests that there are reform elements where they are similar and where they differ. Here four core elements of co-implementation are identified and assessed according to where the respective reform doctrines stand on each: (a) mission clarification; (b) decentralization of decision-making authority; (c) data-driven problem identification and assessment, and (d) innovative problem-solving tactics. These were selected because they appear most relevant to the strategic decisions of first-line supervisors.² Doctrines are essentially theoretical abstractions about how things are supposed to work and not how they actually do so. They are used here to help structure inquiry into how first-line supervision actually operated under these reforms.

One of the key distinguishing features of these reforms is their mission. Compstat’s core mission is to reduce serious crime (McDonald, Greenberg, & Bratton, 2002). Indeed, its progenitors in New York City Police Department saw it as a way of getting back to the essential crime control element of the police (Timoney, 2010). In contrast, community policing encourages a substantial broadening of the police mission to a wider range of objectives which community members help identify as priorities. Crime reduction remains essential, but equally important are reducing fear of crime, responding to quality-of-life problems, and addressing minor problems and disorders (e.g., noise, abandoned cars, public drinking, graffiti; Skogan & Hartnett, 1997, p. 8).

An additional core element of both reforms is the decentralization of decision-making authority to make police organizations more flexible and responsive to local conditions, but so far, these reforms have pursued it to different degrees. Compstat has concentrated on the delegation of decision-making authority to middle managers, primarily at the district/precinct commander level. Community-policing approaches have been quite diverse, but, in general, there has been more interest in decentralizing decision making much further down the organization to the small area or beat level (Maguire, 1997). Opinions vary on whether community-policing responsibilities should be assigned to specialists working independently or in units, or to uniformed patrol, but generally community-policing reformers have stressed the benefits of despecialization (Bayley, 1994, p. 46).

As for data, both reforms emphasize crime analysis as a knowledge base for driving strategic responses to crime and social disorder. Where they differ is Compstat has tended to use a department’s existing information systems (particularly police reports and computer-aided dispatch) and traditional measures of success (e.g., clearance rates and levels of reported crime). In contrast, community policing embraces nontraditional sources of information (e.g., community surveys and beat meetings) and additional performance measures (e.g., citizen fear levels and indicators of physical and social disorder) that are rarely captured in police databases (Peak & Glensor, 1999, pp. 94-96).

Finally, under Compstat and community policing, data are supposed to provide a basis for searching for and implementing innovative solutions. Police are expected to select responses that offer the best prospects of success, not because they are “what we have always done” but because a careful consideration of a number of alternatives showed them the most likely to be effective at reducing crime and disorder problems (Reisig, 2010; Silverman, 1999, pp. 123-124). These solutions may include law enforcement responses, such as arrest, but these should only be selected after a thorough assessment of alternatives.

Compared with the traditional model of policing, the full implementation of these reforms would seem to require significant changes in how first-line supervisors learn about crime and disorder problems and exercise their judgment in ways that are most likely to produce desired results. How, if at all, had first-line supervisors adapted to the major strategic elements of Compstat and community policing?

Method

In 2006, a national mail survey was administered to large municipal and county police agencies with 100 or more sworn officers according to the 2000 Census of State and Local Law Enforcement Agencies (excluding sheriff’s departments). Of the 566 agencies in this sample pool, 355 (63%) responded to the survey. Respondents were asked to what extent their departments had adopted Compstat and/or community policing and what form they took. However, the survey instrument could not give in-depth insights on how this reform worked “on the ground.” Consequently, the survey findings were used to identify seven large (>1,000 sworn), medium (500-999), and small (100-499) police agencies suitable for onsite fieldwork. The largest agency selected was the Los Angeles Police Department, California, followed by the Montgomery County Police Department, Maryland. The two medium agencies were the St. Louis County Police Department, Missouri, and Colorado Springs Police Department, Colorado. The three smallest were Overland Park, Kansas, Marietta, Georgia, and Cape Coral, Florida (see Appendix A for a profile of these departments).

Departments were asked to report on the implementation amount or “dosage” of each reform. For Compstat implementation, questions included to what extent the department concentrated on a single mission and used crime statistics to identify priorities and reallocate resources. Among community-policing items, departments reported on to what extent they gave community groups a say in establishing priorities and delegated authority to patrol officers to initiate projects.

Departments were selected for onsite visits that scored among the highest for both Compstat and community policing implementation and had been operating both reforms for at least 4 years. This helped ensure that visits were conducted at those departments where these reforms were a major part of daily operations and which had been operating long enough to overcome initial implementation problems. The validity problems associated with having a single respondent make broad generalizations about the level of implementation of a specific program are well known (Maguire & Mastrofski, 2000). Indeed, a systematic analysis of this project’s fieldwork data suggested there was considerable variation in how each of these reforms was implemented across sites (Willis, Mastrofski, & Kochel, 2010b). Nonetheless, this assessment also confirmed that each department had either fully or moderately adopted nearly all the key elements of these reforms.

Five-day site visits were conducted between July 2006 and June 2007. The research activities included observing department activities, community meetings, and Compstat meetings at the department and district levels, and interviews. In addition, 2-hr focus groups were conducted with six (on average) first-line supervisors, a many more would have been unwieldy (Krueger, 1994). Unfortunately one of the groups had to be cancelled due to scheduling issues,³ and so the data presented here are from six focus groups.

Given resource constraints, focus groups were chosen as an efficient method of gleaning multiple perspectives on the operational decision-making experiences of first-line supervisors within different agencies (Patton, 2002). Moreover, not only do focus groups allow for the surfacing of individual ideas, they are also a useful way for detecting where differences of opinion and consensus emerge (Krueger, 1994). To elicit thoughtful and in-depth insights on the nature of first-line supervision under these reforms, the department liaison who assisted with scheduling the on-site visit was asked to identify 6 to 10 first-line supervisors who were likely to be knowledgeable, perceptive, and articulate about Compstat and community policing, and who might be willing to participate in a focus group. As a variety of experiences were sought, the only additional stipulation was that the sergeants were not “cherry-picked” based on whether they were likely to express uniformly positive attitudes toward these reforms.

Of the final pool of 34 participants, the majority participants (23) were generalist patrol sergeants, 6 were assigned to community-policing units, 3 were assigned to traffic, and 2 were in administrative positions. The number of patrol officers supervised by generalist patrol sergeants also varied from as few as 5 to as many as 30 (although in this particular case, the sergeant was not in charge of a squad, but was a watch commander). As the focus groups were held on the last morning of the site visit (a Friday), day-shift sergeants were overrepresented in the sample. As the day shift is often regarded as the most desirable shift and generally assigned according to seniority, most of these sergeants would be considered veterans with 80% having upward of 15 years experience.

Given that the sample was small and not randomly selected, these findings are obviously not generalizable and should be interpreted with caution. It might be that those sergeants heard from were selected for reasons other than those that were stipulated or that the sergeants in the focus groups were less willing to adapt to the strategic elements of these reforms than less experienced supervisors who might be more receptive to change. Given these limitations, this research is best thought of as an exploratory study of strategic decision making by veteran first-line supervisors in co-implementing agencies.

An interview guide consisting of open-ended questions helped structure the focus group discussions to facilitate comparisons across sites (Morgan, 1997). Consistent with the literature on focus groups (Krueger, 1994; Morgan & Spanish, 1984), after a brief description about the purpose of the study, the focus group leader asked some introductory questions. Next was a transition to the study’s main questions on how sergeants made decisions about specific crime and disorder problems. These were followed with some more general questions about the challenges of co-implementation and suggestions for how Compstat and community policing could be integrated more closely. Finally, the session ended with a summing up (to ensure an accurate representation of what was said) which also gave participants the opportunity to correct any errors or offer any final insights. In addition to covering a range of topics associated with strategic decision making, the focus group leader sought specificity by directing discussion to concrete examples and depth by way of several follow-up probes (Merton, Lowenthal, & Kendall, 1956; for a copy of the focus group instrument, see Appendix B). All of the focus groups were taped except for one which was not taped due to a technical difficulty. In this case, the focus group leader typed up his notes immediately after it ended to recall as much information as possible.

Each focus group tape was listened to at least twice and general themes were identified. Once this phase of the analysis was complete, the researcher looked for patterns, as well as any differences between them, to be sensitive to variations across sites. As the policing literature has not given much attention to the relationship between Compstat and community policing, this analysis was mostly inductive. Working back and forth between the data and the patterns in responses, the focus was on identifying the clear and consistent themes that emerged from this analysis (Patton, 2002).

The focus group guide provides the structure for the results section below so that sergeants’ responses are not artificially detached from the context in which they arose. It focuses on two key questions: (a) What was the nature of the guidance that first-line supervisors received from their superiors on how to address a crime or neighborhood problem? and (b) What kind of direction did first-line supervisors give to their officers in terms of how these problems should be handled? In what follows, I begin by describing how strategic decision making should influence first-line supervision under Compstat and community according to the reform literature before examining how sergeants said it actually did operate.

Results

Guidance From Superiors to First-Line Supervisors on Crime and Neighborhood Problems

One of Compstat’s principal objectives is to bend the performance of the organization to the chief executive’s will, and to do this by empowering middle managers (or district commanders) to respond to the chief’s direction (Weisburd et al., 2003). Regular Compstat meetings can be a powerful means of clarifying what is most valued by top leadership (reductions in serious crime) and for holding district commanders accountable for accomplishing the organization’s crime control mission. To be consistent with this approach, pressure for results should show up in the kind of direction that first-line supervisors receive from their superiors. As the crime reduction efforts of district commanders are being frequently monitored and assessed, it seems reasonable to assume that they would issue specific directives about how a particular crime or neighborhood problem should be handled effectively and care about whether and how their commands were in fact implemented at the street level.

From the perspective of community policing, which devolves decision-making authority further, one would expect supervisors to receive some direction about which problems were most important to address (and many of these should be concerns identified by community members), but be given more autonomy than under Compstat (Skogan, 2006a).

There was a consensus among the focus group sergeants that the kind of guidance they received from above depended a great deal on a district commander’s particular management style. However, the guidance they were most likely to receive routinely concerned the identification of a specific crime spike, trend, or pattern. On occasion, sergeants might also be told to address a community complaint, but this was less likely. Moreover, it was not customary to be given explicit instructions about how best to respond. One sergeant referred to patrol’s knowledge and experience. He said that when the captain puts out a memo that says there has been a 43% increase in larcenies from autos:

. . . he is saying, O.K. guys, you have one hundred years of experience between the three of ya’ let’s fix it. That is what he is doing: it is not a direct, “this is what we are going to do.” He throws this thing out there. Put your heads together and figure out how we are going to resolve this issue.

In another focus group, a sergeant made a similar point about being left alone:

The directive we get from our commander . . . when he has specific things he wants dealt with, he tells a lieutenant and it is passed down and these are usually things that are generated as a result of a complaint that he has gotten or a crime trend he has seen that he wants to make sure we are handling.

“Handling” could be as simple as the sergeant making sure that “fires were being put out,” or working toward a more specific goal: “We’ve got 32 larcenies from autos, I want to see 18 next week.” In both cases it was up to the sergeant to resolve the issue. The emphasis on addressing crime rather than disorder problems within the context of measurable results is certainly consistent with Compstat. However, the disinclination of district commanders to exert control over how first-line supervisors chose to respond is less consistent with Compstat doctrine, which seeks to harness the organization more tightly to top management’s objectives. Strengthening accountability for results is a key part of Compstat’s design and how it was implemented at most of these sites. Nevertheless, research suggests that in practice Compstat encourages timely responses to crime problems, but top management is less concerned with the quality of those responses (Willis, Mastrofski, & Weisburd, 2007). Allowing sergeants to address crime problems any way they saw fit, is consistent with these previous research findings. Sergeants were expected to respond to the concerns transmitted from above, but district commanders did not provide them with a carefully crafted plan about what to do because this was not a key feature of each department’s Compstat’s program: It was sufficient for sergeants’ to be doing something.

As for community policing, sergeants should be granted considerable discretion about how to mobilize but in ways that are structured to accomplish this reform’s desired goals. The fact that community policing is distinguished by several objectives and approaches would seem to place an additional burden on district commanders to provide their sergeants with manageable choices, including instructions on how to judge which objectives or considerations were most important. Focus group participants did not report receiving this kind of guidance and expectations for community policing appeared much more diffuse than under Compstat (one sergeant exclaimed that he wished someone would tell him how community policing was actually defined). Sergeants’ comments suggested they were generally aware of the need to respond to community concerns, but they received little direction on the more strategic elements of this reform (e.g., problem solving and mobilizing local neighborhood resources) and their specific relationship to an overarching community-policing philosophy. Consequently, sergeants generally understood community policing as their having to demonstrate a commitment to the needs of the community and behaving courteously during individual encounters.

One of the most likely explanations for this rather limited conception was every department’s decision to create specialist community-policing units rather than assigning community-policing responsibilities to all of uniformed patrol. Across sites, these specialist units comprised only a very small proportion of all officers assigned to patrol (approximately 5% to 10% in my assessment based on department records). Freed from answering 911 calls, community-policing officers were expected to respond to a wide range of minor crime problems or quality-of-life issues, thus freeing up patrol officers to respond to calls for service. Separate interviews with community-policing sergeants suggested they were much more familiar with the overall philosophy of this reform, particularly the idea that it was an organizational strategy that went beyond the kind of community relations described by the generalist patrol sergeants.

Although district commanders generally eschewed giving sergeants specific instructions on how they should mobilize, data were fundamentally important to shaping the decision of where and when to mobilize. Most important in this regard were officer reports on individual crime incidents with crime analysts paying particularly close attention to Part I crimes. Information on these data was generally disseminated electronically through reports, spreadsheets, and maps. Sergeants used these to identify which crimes were up or down and where they seemed to be concentrated. For example, one focus group member said:

Our mall has been getting hammered pretty bad with larcenies. And just trying to remind people [her patrol officers] that this is out there, that we are getting hit pretty hard, that they are occurring between 1 and 4 o’clock in the afternoon . . . get descriptions and pretty much set them on their way to go out and fight crime.

As none of these departments had implemented similarly sophisticated data systems to support community policing, sergeants did not mention receiving information that helped them systematically identify community problems, determine priorities, and document results. Consequently, sergeants tended to learn about those issues on a more ad hoc basis, such as through communication with their district commanders and community-policing units, and by asking their patrol officers and individual citizens.

Guidance From First-Line Supervisors to Patrol Officers on Crime and Neighborhood Problems

As first-line supervisors are the “transmission belt” for translating policy into practice, it is important to know how decisions flowed down the chain-of-command (Skogan, 2008b, p. 25). One of the powerful forces shaping Compstat’s inception as a crime-control approach was the concern that many patrol officers were too young and inexperienced to make decisions about how to tackle complex crime problems (Bratton & Knobler, 1998, p. 199). In contrast, community policing recognizes that the vast majority of police work is conducted by the rank and file who are regarded as a vital knowledge resource for rapidly identifying locally defined problems and helping develop local solutions (Skogan, 2006b, p. 38). What was the nature of the guidance that first-line supervisors offered to their patrol officers?

The most common theme that arose was how any form of strategic decision making among the lower ranks was strongly hindered by the demands of each department’s 911 response system. Almost universally, focus group members stated that calls for service undermined their capacity to give strategic guidance to their patrol officers. As one sergeant said, “Unless there is a specific problem that has to be tackled at a particular time in a particular way, it is a matter of getting to deployment assignments when you can and between calls for service.”

In response to a follow-up question, a sergeant in a different department said, “there is only so much you can do; you are tied to calls for service.” Obviously when the calls were about violent crime, there was even more pressure to respond quickly. In one of the larger departments, a participant said,

If you are in black-and-white you are going from violent crime to violent crime, to trying to handle radio calls as quickly as possible, to going to violent crime. You don’t have time to sit and just develop relationships with people all day.

Estimates of how much time the officers they supervised spent answering calls varied (approximately 40% to 50% of their time). Engaging in reactive or preventive patrol while responding to calls to intervene in those individual situations where “something-ought-not-to-be-happening and about which someone had better do something NOW!” are core features of the traditional policing model (Bittner, 1990, p. 249). Both reforms challenge this model by attempting to reallocate resources to those problems which are the most pressing. Rather than devoting substantial time and effort to dealing with all manner of citizen requests as they arise, officers should be targeting specific problems (Skogan et al., 1999, pp. 3, 35). Research in the 1970s and 1980s largely discredited the crime control benefits of this approach, but it remains a mainstay of American policing because citizens want to see patrol cars in their neighborhoods and receive timely service from police when they are summoned (Mastrofski & Willis, 2010).

The finding that Compstat’s emphasis on rapid responses to crime problems that did little to promote creative solutions from district commanders or sergeants was reaffirmed when sergeants were asked to describe what kinds of approaches they expected their officers to adopt in response to crime and neighborhood problems. First-line supervisors did not mention using data to analyze problems and propose creative or innovative solutions. Rather, they wanted their officers to show that they were active or doing “something” to address crime problems and to stay busy. Just as district commanders wanted sergeants to show they were addressing the crime spikes or trends that they had identified, sergeants wanted the same from their officers.

Although it was possible that sergeants would encourage their officers to come up with creative problem-solving solutions, their comments suggested that it was more important that they engaged in activities that could be easily measured such as generating tickets and making arrests. When sergeants were asked what they wanted to see from their officers in terms of addressing specific crime problems, discussion invariably centered on a standard tool kit of law enforcement responses: location-directed patrol, traffic enforcement, and arrest. For example, in response to what he expected his officers to do to address his city’s crime problems, a sergeant replied,

Typically, I mean when you are talking about bar problems, construction area thefts, they don’t demand an enormous amount of creativity in how you approach them . . . you either want to be seen or you don’t want to be seen. If you want to be seen, you want to patrol to increase visibility, do traffic enforcements, do bar checks.

This comment is indicative of those made by other respondents, but the context in which it is mentioned is also interesting because it reveals that these departments’ focus on productivity to assess individual officer performance did little to promote problem solving. One of the challenges to community policing is that, unlike calls for service and reported crime, its accomplishments are not easily measured (Skogan, 2008a). In the focus groups, sergeants tended to emphasize traditional indicators that were easily captured by existing databases as measures of officer performance. Community-policing approaches often require that officers make artful judgments (e.g., resolving a long-standing neighborhood dispute), but, as one of the community-policing sergeants said, these are rarely documented and recorded and so are undervalued.

As mentioned earlier, from a community-policing perspective, traditional information systems of the kind used by Compstat are in many respects inadequate for the purpose of identifying and learning about many issues that concern citizens most. One means of learning about these and mobilizing accordingly is for officers to attend community meetings and work with local residents, business owners, and other neighborhood organizations to focus proactively and creatively on problems (Skogan, 2006a).

At none of the sites were patrol sergeants and their officers expected to attend regular community meetings and collaborate with community members or organizations to address neighborhood problems: this was the domain of community-policing units. As a result, first-line supervisors may have encouraged their officers to surface and respond to community problems, but this was only likely to occur on a desultory rather than a routine basis. Innovative problem solving was undermined by two additional factors: sergeants’ many other responsibilities and the reluctance of each department to allow first-line supervisors to make key decisions about reallocating patrol resources.

Sergeants had to juggle their responsibility for ensuring public safety with myriad supervisory and administrative tasks. These included approving reports, cultivating morale, and helping officers advance their careers. These essential functions of first-line supervisors have not been lost on scholars or on sergeants themselves (the “boss” in Van Maanen’s [1983] evocative terminology; p. 275). There was consensus among participants that sergeants were indispensable, with one capturing this sentiment when he said, “a sergeant is probably the most important position in the whole department.” At the same time, these administrative burdens meant sergeants did not have time to innovate by drawing on knowledge gained in other departments and innovations in theory and research about crime control and prevention.

Another essential element of the problem-solving process is increasing a department’s capacity to move resources to where a problem is and to change or disrupt department routines to do this. Although sergeants acknowledged that they had significant autonomy in deciding how to respond to problems in their beats, they also mentioned that this was not as extensive when it came to the reallocation of resources. Although they may have been allowed to approve over-time, any “special” or unusual requests (e.g., putting an officer in plainclothes, asking officers to work on their days’ off), or those that demanded a change to how resources were routinely allocated (e.g., coming up with an operational plan that required freeing several officers from answering calls for service), required approval from the district commander.

In summary, at these sites a variety of factors conspired to limit the capacity of sergeants to engage in the kind of innovative problem solving that is regarded as a hallmark of both Compstat and community policing. Some of these were related to features of the programs as they had been implemented, such as Compstat’s focus on rapid responses to crime and the absence of structured problem-solving partnerships with community members, but other factors were features of general police organization. Across sites, the traditional 911 service and existing performance systems did little to support innovation.

Discussion

According to their supporters, both Compstat and community policing promise to transform police organizations radically, particularly in terms of making them respond strategically to crime and disorder problems. Given the central role that sergeants play in implementing these reforms at the street level, it would seem reasonable to expect to see significant changes at this rank. These focus groups, however, indicated a pattern of practices that did not readily fit with the idealized models of either Compstat or community policing. As these sergeants were purposively selected within departments that had assigned primary community-policing responsibilities to specialist units, these sites cannot represent the experiences of all co-implementing departments. However, the patterns described here provide some valuable insights into how these specific reforms operated and into police reform more generally.

The unevenness of co-implementation can be explained by each department continuing to stress features of police organizations that are most consistent with goals that have long been embraced by top police managers: fighting crime, centralizing decision-making authority, and responding to calls for service (National Research Council, 2004). Those elements of these reforms that represented the greatest departure from past management and supervisory ideals—community involvement in the production of police priorities and of responses to crime and neighborhood problems, devolving decision-making authority to the rank and file, and innovative problem solving—remained the least developed.

The most noticeable influence of these reforms on sergeants’ decision making was guidance and the use of crime data for the rapid identification of serious crime problems. Sergeants were most likely to receive instruction from their superiors on crime trends and patterns. Moreover, sergeants also relied on these Compstat data to make decisions about where and when to focus patrol resources. Similar data systems were not in place for incidents of minor crime and social disorders, no doubt hampering sergeants’ ability to conceive of community policing in more strategic terms, or as an approach designed to “address the causes and reduce the fear of crime and social disorder” (Scheider, Chapman, & Schapiro, 2009). Rather, sergeants tended to think of community policing as synonymous with the kind of “service-style” policing first identified by James Q. Wilson in the late 1960s (Wilson, 1968).

There was not an opportunity to observe these departments before Compstat and community policing were implemented, but based on respondents’ comments during these site visits, these changes were significant. Crime analysis had come to play an integral role in police operations, particularly when it came to allocating resources to hot spots. Research shows that focusing patrol on areas where crime is concentrated can have significant crime reduction benefits (Braga & Weisburd, 2010). In addition, increasing the responsiveness of the police to the communities they serve has long been a core goal of community policing, and sergeants seemed to value the kind of customer service approach that this evokes (Mastrofski, 1999).

Despite the incorporation of these reform features into sergeants’ decision-making routines, co-implementation had fallen short of its strategic promise. The reason for this is a mainstay of the literature on police organizational change (Maguire, 1997; Mastrofski, 2006; Mastfroski & Willis, 2010; Sadd & Grinc, 1994; Skogan, 2006a). Significant change requires that police organizations go beyond programs and activities, by putting new organizational structures in place (Skogan, 2006a, p. 29). The sites here, however, were either unwilling or unable to do so, preferring to implement a specialist community-policing model which insulated them from making more radical changes. Envisioning what form these changes might have taken and their implications for first-line supervision is a useful way of assessing the lack of change that actually did take place. It also helps identify policy-relevant strategies for how first-line supervision could be harnessed more directly to the goals of these reforms. A recent article making recommendations for integrating Compstat and community policing is instructive here, because its suggestions are readily adapted to first-line supervision (Willis, Mastrofski, & Kochel, 2010a).

Creating performance measures at the organizational level to systematically identify and prioritize community concerns could help reinforce the fundamental importance of community policing’s strategic mission to key decision makers, including sergeants. So, for example, departments might require that community-identified problems are routinely reported at Compstat meetings to focus the department on community-policing priorities (a practice we did not observe). Prioritizing and reporting community-policing concerns in this way would help focus and clarify their essential importance to the organization’s core mission.

This approach could be further supported by an individual performance evaluation system that put less emphasis on traditional police responses and provides more incentives to engage in problem solving. Requiring sergeants to assess regularly the level of creativity of their officers in response to persistent crime and disorder problems, especially those identified by community members and not just those identified by the police, could help promote the kind of innovative thinking called for by both reforms.

Moreover, assigning patrol officers to permanent beat teams supervised by patrol sergeants and delegating responsibilities to these teams and not to community-policing units might increase accountability for community policing and problem solving across the organization. These teams could be made responsible for meeting with community members regularly and working with them to identify their most pressing concerns and figuring out the best ways to respond. Similar to Compstat’s accountability mechanism, sergeants would experience accountability by having to demonstrate their understanding of the community’s problems and the progress they were making in front of an audience of local residents and business owners directly invested in operational outcomes.

To be most effective, this would require that sergeants and patrol officers (and possibly community members) receive adequate training in the basic skills of problem solving. This would mean sergeants and their beat teams learning (a) when and how to identify a problem; (b) how to make reasonable suggestions for addressing it based on knowledge of the beat, conversations with stakeholders, and input from crime analysts; and (c) how to report on results of problem-solving efforts at community meetings. As part of this process, officers could be trained in the well-known SARA (scanning, analysis, response, and assessment) and crime triangle problem-solving models and could be introduced to other problem-solving resources, such as POP (problem-oriented policing) guides.

Perhaps most important for the successful realization of the strategic vision articulated by reformers would have been for departments to reorganize how daily work was assigned and managed by making changes to their 911 systems. Chicago, for example, assigns a team of officers to handle low priority calls thereby freeing up other officers that first-line supervisors can then reassign to tackle persistent crime or disorder problems (Skogan, 2006a, pp. 56-59). Such changes are costly and raise many other challenges, only one of which can be addressed here based on feedback from the focus groups. Would patrol sergeants have been interested in moving away from the specialist approach that characterized co-implementation in their agencies? Given that recent surveys of law enforcement organizations in the United States suggest that such a specialization approach is very popular, this is an important issue (Reaves, 2010, p. 27).

The sergeants in the focus groups were divided over whether primary responsibility for the implementation of community policing should fall to uniformed patrol or specialist officers. Generalist patrol officers were more likely than sergeants assigned to specialist community-policing units to support a specialist approach. Thus, we heard the following comments:

The most effective way of community policing that we have ever done, is to generate specialist units to deal with the problem; when there are 1 or 2 folks available, they join in to be part of it . . . If you have one or two officers free, do you feel like going off and tackling a problem by yourself . . . no . . . but if you have a group of officers, you kind of join in.

And,

Retain traditional policing—clear chain-of-command, clear designation of supervision—couple that with specialized units to focus on problems: undercover teams, specialist assignment teams. (Emphasis in original)

Alternatively, a sergeant assigned to a community-policing unit in another department, one that had disbanded a community-policing approach based on geographic decentralization, said: “What is lost is COPS. If I am deploying people out differently each day, there is no tie-in and development of relationships within the community, and ownership and even accountability to some degree.”

If these comments are any indication of the sentiments shared by others in these departments and further afield, many officers’ hearts and minds would have to be won for the successful implementation of a less specialized community-policing approach.

Conclusion

In closing, it might be useful to speculate whether these focus groups reveal a larger trend in U.S. policing where Compstat and community policing, despite the warnings of their supporters, are defined by their activities rather than as a fundamental transformation in how police departments operate (Silverman, 2006; Skogan, 2006a). Future research might examine this issue further, perhaps comparing first-line supervision in co-implementing departments with Compstat or community-policing-only agencies. If what was found in these focus groups characterizes U.S. policing more generally, reformers might consider experimenting with the recommendations described above that could strengthen strategic decision making by first-line supervisors under both reforms. Of course, until these are tested there is no way to know if they would work as intended, but given that much hope and effort has been invested in these reforms, these suggestions might be worth considering. What is clearer is that in the absence of changes designed to work to the mutual benefit of Compstat and community policing, it seems likely that the strategic potential of co-implementation will remain more contingent upon the will and skill of individual sergeants than reformers might realize.

Footnotes

Appendix

Appendix B

Focus Group Guide

Introduction (Who we are, why we are visiting).

Goal: Learn from you—learn about how you make decisions and how you and your officers spend your time.

Using a focus group because this allows us to hear a variety of perspectives on decision making within the department—it also allows us to hear many opinions in a short time.

Logistics: Informed consent—voluntary, no benefit, won’t be identifying you in the final report, and will be taping.

Helpful ground rules:

Please contribute, even if your perspective differs from others.

There are no right answers here, so please feel free to speak your mind.

It is helpful for me if you can provide specific examples of your experiences.

My role is to facilitate discussion—keep it on track and encouraging everyone to speak—I’ve got one shot at this, so my primary interest is getting as much information from you as possible.

To get things moving—would you complete the name tents by writing your name on one side and then writing your contact information on the back (shift, current assignment, how many officers you currently supervise, contact information).

Once you have done that perhaps we could go round the table and you could:

Introduce yourself

Tell us your shift

No. of officers that work under your direct supervision

How long you’ve been with the PD

% of time on average shift that your estimate your officers spend on calls for service.

I know you have numerous administrative tasks like reading and signing reports, attending meetings, and giving advice to patrol officers but here I am interested in how you spend your time when dealing with crime and neighborhood/quality-of-life problems.

Perhaps you could think about the last time you got guidance or direction from a superior on how to deal with a crime or neighborhood problem. What was the nature of that guidance?

What do you think caused your superior to give you guidance in this case?

How detailed was the guidance? (Priority or degree of importance? Time? Place? Location?)

Over say, the past week, perhaps you could talk about a specific crime problem that you have been working on. How was this identified? What kinds of things have you been doing to deal with it?

What kinds of activities have you asked your officers to do?

How often?

What spurs?

How detailed? (Priority/degree of importance? Time? Place? Location? How to respond?)

When not responding to calls, what kinds of things do you expect your officers to be doing in response to crime and neighborhood problems?

To what extent is crime analysis helpful in how you deal with crime problems?

In your mind, how well do CS and CP work together? What are some of the challenges that have arisen? What about the benefits?

Finally, what recommendations do you have for improving how these programs work together?

To finish let me summarize some of the main points—Is that accurate? Is there anything you would like to add or change?

Authors’ Note

The opinions contained herein are those of the author and do not necessarily represent the official positions or policies of the U.S. Department of Justice. References to specific agencies, companies, products, or services should not be considered an endorsement by the author(s) or the U.S. Department of Justice. Rather, the references are illustrations to supplement discussion of the issues.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Office of Community-Oriented Policing Services, U.S. Department of Justice [Cooperative Agreement Number 2005-CK-WX-K003].

Notes

Author Biography

James J. Willis is an assistant professor in the Department of Criminology, Law & Society at George Mason University. His research interests include police organizational reform and police decision making.

References

Allen

(1982). Police supervision on the street: An analysis of supervisor/officer interaction during the shift. Journal of Criminal Justice, 10, 91-109.

Bayley

(1994). Police for the future. New York, NY: Oxford University Press.

Bittner

(1990). Florence Nightingale in pursuit of Willie Sutton: A theory of police. In Aspects of police work (pp. 233-268). Boston, MA: Northeastern University Press.

Braga

Weisburd

(2010). Policing problem places: Crime hot spots and effective crime prevention. New York, NY: Oxford University Press.

Bratton

Knobler

(1998). Turnaround: How America’s top cop reversed the crime epidemic. New York, NY: Random House.

Clarke

R. V.

(2004). Defining police strategies: Problem solving, problem-oriented policing, and community-oriented policing. In Thurman

Q. C.

Zhao

(Eds.), Contemporary policing: Controversies, challenges, and solutions (pp. 18-38). Los Angeles, CA: Roxbury.

Cordner

Biebel

(2005). Problem-oriented policing in practice. Criminology & Public Policy, 4, 155-180.

Dabney

(2010). Observations regarding key operational realities in a Compstat model. Justice Quarterly, 27(1), 28-51.

Eck

Spelman

(1987). Problem-solving: Problem-oriented policing in Newport News. Washington, DC: Police Executive Research Forum.

10.

Engel

(2002). Patrol officer supervision in the community policing era. Journal of Criminal Justice, 30, 51-64.

11.

Eterno

J. A.

Silverman

E. B.

(2006). The NYPD’s Compstat: Compare statistics or compose statistics. Policing: An International Journal of Police Science & Management, 12, 426-449.

12.

Federal Bureau of Investigation. (2006). Crime in America. Retrieved from http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2006/crime2006

13.

Goldstein

(1987). Improving policing: A problem-oriented approach. Crime & Delinquency, 25, 236-58l.

14.

Henry

V. E.

(2002). The COMPSTAT paradigm: Management accountability in policing, business, and the public sector. Flushing, NY: Looseleaf.

15.

Kelling

G. L.

Moore

M. H.

(1988). From political to reform to community: The evolving strategy of police. In Greene

J. R.

Mastrofski

S. D.

(Eds.), Community policing: Rhetoric or reality (pp. 3-26) New York, NY: Praeger.

16.

Kim

W. C.

Mauborgne

(2003, April). Tipping point leadership. Harvard Business Review, 1-12. Retrieved from https://archive.harvardbusiness.org/cla/web/pl/product.seam?c=14017&i=14019&cs=61e2b53cc3609b79d96c08525d9b1a19

17.

Krueger

(1994). Focus groups: A practical guide for applied research. Thousand Oaks, CA: SAGE.

18.

Maguire

E. R.

(1997). Structural change in large municipal police organizations during the community policing era. Justice Quarterly, 14, 547-576.

19.

Maguire

E. R.

Mastrofski

S. D.

(2000). Patterns of community policing in the United States. Police Quarterly, 3(1), 4-45.

20.

Manning

P. K.

(1977). Police work: The social organization of policing. Cambridge, MA: MIT Press.

21.

Maple

Mitchell

(1999). The crime fighter: Putting the bad guys out of business. New York, NY: Doubleday.

22.

Mastrofski

S. D.

(1999). Ideas in American policing: Policing for people. Washington, DC: Police Foundation.

23.

Mastrofski

S. D.

(2006). Community policing: A skeptical view in Police Innovation: Contrasting Perspectives (pp. 44-73). Cambridge, UK: Cambridge University Press.

24.

Mastrofski

S. D.

Willis

J. J.

(2010). Police organization continuity and change: Into the twenty-first century. In Tonry

(Ed.), Crime and justice: A review of research (pp. 55-144). Oxford, UK: Oxford University Press.

25.

McDonald

Greenberg

Bratton

(2002). Managing police operations: Implementing the New York crime control model—Compstat. Belmont, CA: Wadsworth.

26.

Merton

R. K.

Lowenthal

M. F.

Kendall

P. L.

(1956). The focused interview: A manual of problems and procedures. New York, NY: Free Press.

27.

Moore

M. H.

(1992). Problem solving and community policing. In Tonry

Morris

(Eds.), Modern policing (pp. 95-158). Chicago, IL: University of Chicago Press.

28.

Morgan

D. L.

(1997). Focus groups as qualitative research. Thousand Oaks, CA: SAGE.

29.

Morgan

D. L.

Spanish

M. T.

(1984). Focus groups: A new tool for qualitative research. Qualitative Sociology, 7, 253-270.

30.

National Research Council. (2004). Fairness and effectiveness in policing: The evidence. Committee to Review Research on Police Policy and Practices. In Skogan

Frydll

(Eds.), Committee on Law and Justice, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press.

31.

Patton

M. Q.

(2002). Qualitative research and evaluation methods. New York, NY: SAGE.

32.

Peak

Glensor

R. W.

(1999). Community Policing and Problem Solving: Strategies and Practices. Upper Saddle River, NJ: Prentice Hall.

33.

Ratcliffe

(2002). Intelligence-led policing and the problems of turning rhetoric into practice. Policing & Society, 12, 53-66.

34.

Reaves

B. A.

(2010). Local police departments, 2007. Washington, DC: Bureau of Justice Statistics.

35.

Reisig

(2010). Community and problem-oriented policing. In Tonry

(Ed.), Crime and justice: A review of research (pp. 1-53). Chicago, IL: University of Chicago Press.

36.

Rosenbaum

(Ed.). (1994). The challenge of community policing: Testing the promises. Newbury Park, CA: SAGE.

37.

Rubinstein

(1973). City police. New York, NY: Farrar, Straus, and Giroux.

38.

Sadd

Grinc

(1994). Innovative neighborhood oriented policing: An evaluation of community policing programs in eight cities. In Rosenbaum

D. P.

(Ed.), The challenge of community policing: Testing the promises (pp. 27-52). Thousand Oaks, CA: SAGE.

39.

Scheider

Chapman

Schapiro

(2009). Toward the unification of police innovations under community policing. Policing: An International Journal of Police Strategies & Management, 32, 694-718.

40.

Silverman

E. B.

(1999). NYPD battles crime: Innovative strategies in policing. Boston, MA: Northeastern University Press.

41.

Silverman

E. B.

(2006). Compstat’s innovation. In Weisburd

Braga

(Eds.), Police innovation: Contrasting perspectives (pp. 267-283). Cambridge, UK: Cambridge University Press.

42.

Silverman

E. B.

O’Connell

(1999). Organizational change and decision making in the New York City Police Department: A case study. International Journal of Public Administration, 22, 217-259.

43.

Skogan

W. G.

(2006a). Police and community in Chicago: A tale of three cities. New York, NY: Oxford University Press.

44.

Skogan

W. G.

(2006b). The Promise of community policing. In Weisburd

Braga

(Eds.), Police innovation: Contrasting perspectives (pp. 27-43). Cambridge, UK: Cambridge University Press.

45.

Skogan

W. G.

(2008a). An overview of community policing: Origins, concepts, and implementation. In Williamson

(Ed.), The handbook of knowledge-based policing: Current conceptions and future directions (pp. 43-57). Chichester, UK: John Wiley.

46.

Skogan

W. G.

(2008b). Why reforms fail. Policing & Society, 18, 23-34.

47.

Timoney

J. F.

(2010). From beat cop to top cop: A tale of three Cities. Philadelphia, PA: University of Pennsylvania Press.

48.

Skogan

W. G.

(Ed.). (2004). Community policing: Can it work? Belmont, CA: Wadsworth.

49.

Skogan

W. G.

Hartnett

S. M.

(1997). Community policing: Chicago style. New York, NY: Oxford University Press.

50.

Skogan

W. G.

Hartnett

S. M.

DuBois

Comey

J. T.

Kaiser

Lovig

(1999). On the beat: Police and community problem solving. Boulder, CO: Westview.

51.

Trojanowicz

(1980). The environment of the first-line police supervisor. Englewood Cliffs, NJ: Prentice Hall.

52.

Trojanowicz

Bucqueroux

(1990). Community policing: A contemporary perspective. Cincinnati, OH: Anderson.

53.

U.S. Census Bureau (2006). www.census.gov

54.

Van Maanen

(1983). The boss: First-line supervision in an American police agency. In Punch

(Ed.), Control in the police organization (pp. 275-317). Cambridge, MA: MIT Press.

55.

Weisburd

Braga

(Eds.). (2006). Introduction. Police innovation: Contrasting perspectives (pp. 1-23). Cambridge, UK: Cambridge University Press.

56.

Weisburd

Mastrofski

S. D.

McNally

A. M.

Greenspan

Willis

J. J.

(2003). Reforming to preserve: Compstat and strategic problem solving in American policing. Criminology & Public Policy, 2, 421-456.

57.

Weisburd

McElroy

(1988). Enacting the CPO role: Findings from the New York City Pilot Program in community policing. In Greene

J. R.

Mastrofski

S. D.

(Eds.), Community policing: Rhetoric or reality (pp. 89-102). New York, NY: Praeger.

58.

Willis

J. J.

Kochel

T. R.

Mastrofski

S. D.

(2010). The co-implementation of Compstat and community policing: A national assessment (Final report). Washington, DC: U.S. Department of Justice, Office of Community Oriented Policing Services.

59.

Willis

J. J.

Mastrofski

S. D.

Kochel

T. R.

(2010a). Recommendations for integrating Compstat and community policing. Policing: A Journal of Policy and Practice, 4, 182-193.

60.

Willis

J. J.

Mastrofski

S. D.

Kochel

T. R.

(2010b). The co-implementation of Compstat and community policing. Journal of Criminal Justice, 38, 969-980.

61.

Willis

J. J.

Mastrofski

S. D.

Weisburd

(2007). Making sense of COMPSTAT: A theory-based analysis of organizational change in three police departments. Law & Society Review, 41, 147-188.

62.

Wilson

J. Q.

(1968). Varieties of police behavior. Cambridge, MA: Harvard University Press.