U.S. Department of Justice Office of Community Oriented Policing Services Image of United States Department of Justice seal COPS Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared for Office of Community Oriented Policing Services 1100 Vermont Ave., N.W. Washington, D.C. 20005 www.cops.usdo j.gov Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha 2003 This report was funded by Grant #2000-CK-WX-K124 from the Office of Community Oriented Policing Services of the U.S. Department of Justice. The views expressed in this report do not necessarily reflect to views or the official policies of the U.S. Department of Justice. ISBN# 1-932582-26-6 TableofContents Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .i Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .iii Chapter 1. Introduction to Early Intervention Systems . . . . . . . . . . . . . .3 Chapter 2. The Components of an Early Intervention System . . . . . . . . . . . .27 Chapter 3. Background : The Origins and Development of Early Intervention Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 Chapter 4. Lessons From the Field : Police Managers' Perceptions and Experiences with Early Intervention Systems . . . . . . . . . . . . . . . . . . . . .77 Chapter 5. Planning and Implementing an EI System . . . . . . . . . . . . . . . . . .99 Chapter 6. Policy and Research Needs for Early Intervention Systems . . . . .105 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123 ExecutiveSummary • Early Intervention (EI) systems are an effective mechanism for enhancing accountability within law enforcement agencies. • Police managers experienced with EI systems report that they have a positive impact on officer performance and that they strengthen supervision. • EI systems are extremely complex administrative tools that require close attention by police managers. • In creating an EI system, careful decisions need to be made about the number of performance indicators to be used and the thresholds to be used to identify potential problem officers. • For an EI system to be effective, sergeants and other supervisors need extensive training about their new responsibilities. • EI systems are consistent with the goals of Community Policing and can help to improve police-community relations problems. • Careful planning is essential in the development of an effective EI system. The key to the planning process is the involvement of officers from all ranks of the department. • EI systems are a "work in progress." Existing systems need to engage in a process of continuous review and fine-tuning. • This report focuses on EI in larger agencies: the exact size and scope of EI systems for small and medium-sized law enforcement agencies is not known at this time. • The law enforcement profession needs to sponsor continuing policy development related to EI systems. • Additional research is required on the impact and effectiveness of EI systems. Acknowledgments This report was supported by Grant #2000-CK-WX-K124 from the U.S. Department of Justice Office of Community Oriented Policing Services (COPS). The views expressed in this report do not necessarily reflect the views or official policies of the U.S. Department of Justice. I would like to express my special thanks to Dr. Ellen Scrivner of the COPS Office for her support of this effort and for her on-going commitment to the issue of police accountability. Pam Cammarata picked up the project after Dr. Scrivner left the COPS Office and has been very supportive of it over the course of the last year. Professor Geoffrey P. Alpert of the University of South Carolina served as a major consultant to this project, and I would like to express my thanks for all of his help. Dawn Irlbeck, my Research Assistant at the University of Nebraska at Omaha, made a tremendous contribution to this report. She had major responsibilities for developing and administering the PERF survey and for analyzing the results that are presented here in Chapter Four. She also proofread the final report and saved me many embarrassments. I would like to thank her for all of her work. Chapter Two and the Appendix report the findings of a survey administered in collaboration with DiscoverWhy and the Police Executive Research Forum (PERF). I would like to thank Dr. Lorie Fridell, Research Director at PERF, and both Brad Ashford and Linda Ordoukhanian at DiscoverWhy for agreeing to collaborate on this survey. This project also involved a series of workshops hosted by four Regional Community Policing Institutes (RCPIs) across the country. This included the RCPIs in Tampa, Florida; Baltimore, Maryland; Chicago, Illinois; and Sacramento, California. I would like to thank the directors and staff of those agencies for their assistance. Sergeant Ed Hudak of the Coral Gables (FL) Police Department assisted Professor Alpert and myself in those efforts. The author of this report has had the benefit of tremendous cooperation from officials at both the Los Angeles County Sheriff's Department (LASD) and the Phoenix Police Department. At the LASD I would like to thank Commander Neal Tyler and Commander William J. McSweeney. At the Phoenix Police Department I would like to thank Lieutenant Dan Olsen, Detective Toye Nash, and Sergeant Ron Snodgrass. The Phoenix Police Department hosted the EI System State of the Art Conference in February 2003. I would like to thank Chief Harold Hurtt, Lieutenant Olsen, Detective Nash, and Sergeant Snodgrass for all of their help with that valuable event. The author of this report provided technical assistance to the Douglas County (NE) Department of Corrections in developing its Performance Enhancement Program (PEP) beginning in the year 2000. That experience provided valuable insights into one agency's planning process. Chapter One Introduction to Early Intervention Systems The Concept of Early Intervention . . . . . . . . . . . . . . . . . . . . . . . .3 A Word About Terminology. . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 ARecommended Best Practice in Police Accountability . . . . . . . . . . . . . . .8 EI Systems and Other Best Practices. . . . . . . . . . . . . . . . . . . . . . .10 Goals and Impacts of EI Systems. . . . . . . . . . . . . . . . . . . . . . . . .10 Individual Officers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Supervisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 The Department . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 Early Intervention and Community Policing. . . . . . . . . . . . . . . . . . . .16 Early Intervention and Problem-Oriented Policing . . . . . . . . . . . . . . . .17 Early Intervention and COMPSTAT. . . . . . . . . . . . . . . . . . . . . . . . .17 Early Intervention and Risk Management . . . . . . . . . . . . . . . . . . . . .18 Early Intervention and Police-Community Relations . . . . . . . . . . . . . . .18 Early Intervention and Racial Profiling. . . . . . . . . . . . . . . . . . . . .19 One Size Does Not Fit All Departments. . . . . . . . . . . . . . . . . . . . . .20 AWork in Progress. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 Introduction to Early Intervention Systems The Concept of Early Intervention An Early Intervention (EI) system is a data-based management tool designed to identify officers whose performance exhibits problems, and then to provide interventions, usually counseling or training, to correct those performance problems. EI systems have emerged as an important mechanism for ensuring police accountability.1 EI systems do more than just focus on a few problem officers. As a 1989 report by the International Association of Chiefs of Police (IACP) explains, an EI system is "a proactive management tool useful for identifying a wide range of problems [and] not just a system to focus on problem officers."2 In an EI system, performance data are entered into a computerized database. These data include departmental use-of-force reports, citizen complaints, officer involvement in civil litigation, resisting arrest charges, and other performance indicators. Some current EI systems use a dozen or more performance indicators while others use a smaller number. An EI system is early in the sense that it helps to identify officer performance problems that do not warrant formal disciplinary action but suggest that an officer is having problems dealing with citizens. The major contribution of an EI system is its capacity to spot patterns of performance and to intervene before problems lead to a serious incident such as a lawsuit, a citizen complaint over excessive force, or some other public crisis involving the department. An EI system warns an officer to the extent that it sends an informal but nonetheless clear message that his or her performance needs improvement. An EI system is officially separate from the formal disciplinary system. It is designed to help officers improve their performance through counseling, training, or coaching. No record of participation in an EI program is placed in an officer's personnel file, although a separate record of participation is usually maintained by the internal affairs or professional standards unit. One or more of the incidents identified by an EI system may warrant formal disciplinary action that is officially recorded, but identification by the EI system remains separate from the disciplinary process. EI systems are data-driven mechanisms of accountability, providing systematic data as a basis for performance evaluations of officers. They differ from traditional performance review systems that rely heavily on subjective assessments such as "works well with people" or "demonstrates initiative."3 The database can identify specific areas of performance that need correcting (e.g., a pattern of citizen complaints alleging rudeness). Bob Stewart, former Executive Director of the National Organization of Black Law Enforcement Executives (NOBLE), says that "If I could choose only one accountability mechanism, it would be an early intervention system." Stewart has seen police performance problems from several perspectives: he was a police officer in Washington, DC, the chief of police in Ormond, Florida, and is now a consultant to communities and law enforcement agencies around the country. Based on this long experience, he now regards EI systems as the single most valuable tool for achieving accountability, primarily because of their capacity to monitor a wide range of officer activities and to spot performance problems at an early stage.4 EI systems are retrospective performance reviews. They do not attempt to predict officer performance based on background characteristics or other factors, rather they indicate that current performance levels, while not warranting disciplinary action, still warrant improvement. Past efforts to develop a methodology for predicting which applicants for police employment will perform well and which are not likely to perform well have not proven successful.5 An EI system provides a basis for counseling or warning an officer that his or her performance needs to improve in the future, and in successful cases, for documenting that the improvement has occurred. EI systems are designed to help officers improve their performance. The names of several EI systems reflect this orientation. The New Jersey State Police EI system is known as the Management Awareness Program (MAP), indicating its purpose of assisting management. In this respect, EI systems represent a significant departure from traditional police disciplinary practices. Law enforcement agencies have been punishment-oriented bureaucracies, with innumerable rules and regulations that can be used to punish an officer, but with few procedures for either rewarding good conduct6 or helping officers with problems. Apart from employee assistance programs (EAP) designed to address substance abuse or family problems, police departments have done relatively little in a formal way to correct problem behavior.7 An EI system helps to identify specific performance problems that need to be addressed (e.g., a tendency toward verbal abusiveness, frequent charges of resisting arrest, etc.). EI systems represent a problem-solving approach to officer performance. The problem in this instance involves questionable officer performance. The problem-solving involves identifying officers in need of assistance and providing that assistance through counseling or training. (The EI problem-solving process is discussed in detail later in this chapter.) As one commander with EI system experience explained, his agency's EI system "provides a way for the department to provide non-disciplinary direction and training before the officer becomes a liability to citizens, the department, and him/herself" (See Chapter Four). EI systems are very similar to COMPSTAT programs. COMPSTAT is one of the most important innovations in police management. It is a data-based system designed to help law enforcement agencies respond effectively to crime and disorder and to hold their managers accountable for their performance.8 Both EI systems and COMPSTAT programs rely on the analysis of systematic and timely data–in the case of COMPSTAT the data involve crime and disorder, with EI systems the data involve individual officer performance. EI systems are consistent with the basic principles of personnel management and human resource development.9 Employers recruit, select, and train employees to effectively serve the goals and objectives of their organizations. Effective personnel management assumes that employee performance is assessed and evaluated on a regular basis, and that the organization takes steps to correct unsatisfactory performance. Contrary to the expectations of many people, EI systems have not encountered significant opposition from police unions representing rank and file officers.10 To be sure, there has usually been grumbling and fear of the unknown, but in practice unions have not succeeded in blocking the operation of an EI system once it is in place. The police managers' survey reported in Chapter Four of this report found that only 16 percent of managers had encountered serious opposition from police unions. Potential opposition from unions is best overcome by involving union representatives in the planning of an EI system (see Chapter Five). EI systems are consistent with the goals of community oriented policing (COP). Community policing creates demands for responsiveness to community residents and in the process new measures of police performance. The traditional measures of the crime rate and the clearance rate are no longer adequate. EI systems have the capacity to quickly document the performance of officers who are not effectively serving the community and provide a basis for remedial action by the department. The available evidence indicates that EI systems are successful in achieving their goals of reducing officer misconduct. An NIJ study of EI systems in three police departments found significant reductions in use-of-force and citizen complaints among officers following EI intervention.11 Commanders with EI system experience, meanwhile, are able to report specific examples of individual officers whose performance improved as a result of intervention. One commander, for example, has described an officer who had on-the-street problems because of an excessive fear of being struck in the face. Intervention counseling identified the problem and retraining in tactics helped the officer overcome the problem. Commanders report many other such success stories. In the survey of managers which is summarized in Chapter Four of this report, about half (49 percent) reported that the system has had a positive impact on the on-the-street performance of their officers, while almost a third (28 percent) reported a mixed impact. No commanders reported a negative impact. Chapter Three of this report discusses in detail the research findings regarding the effectiveness of EI systems. Chapter Four, meanwhile, reports the findings of a survey of police managers with EI system experience. Generally, managers have had very positive experiences with the impact of their systems of supervision and accountability. BOX 1.1 EI Systems: An Overview • Data-based management information system. • Capacity for identifying and correcting performance problems. • Recommended police accountability "best practice" • Separate from the formal disciplinary system • Consistent with the goals of Ccommunity Ppolicing • Consistent with the process of Pproblem-Ooriented Ppolicing • Rely on systematic and timely data • Consistent with the process of COMPSTAT • Careful planning is needed for development and implementation A Word About Terminology Early Intervention This report uses the term early intervention. It does not use the more widely used term early warning. Early warning has a negative connotation, suggesting that the system is primarily oriented toward discipline. One department with a comprehensive EI system found through interviews with officers that they did not like the phrase early warning because of its "big brother" connotation. EI systems are evolving in the direction of more comprehensive personnel assessment systems, for the purpose of examining a broad range of performance issues. For this reason, the more positive early intervention term is used here. Other terms are also appropriate. These include Personnel Performance Index, as used by the Los Angeles County Sheriff's Department, or the Personnel Assessment System (PAS), as used by the Phoenix Police Department. Officers With Performance Problems This report uses the term officers with performance problems rather than the commonly used term problem officers. The latter terms unfairly labels officers and suggests that their performance cannot change. The term officers with performance problems focuses on behavior without labeling an officer and conveys the message that performance can improve. A Recommended Best Practice in Police Accountability EI systems have been recommended by a wide range of organizations as a best practice in police accountability. • A January 2001 report by the U.S. Department of Justice (2001), Principles for Promoting Police Integrity, includes EI systems among its recommended best practices.12 • The Commission on Accreditation for Law Enforcement Agencies (CALEA, 2001) has adopted Standard 35.1.15 mandating EI systems for large agencies. The CALEA Standard states that: A comprehensive Personnel Early Warning System is an essential component of good discipline in a well-managed law enforcement agency. The early identification of potential problem employees and a menu of remedial actions can increase agency accountability and offer employees a better opportunity to meet the agency's values and mission statement.13 The CALEA Standard warns that a department should not be faced with a situation where an officer is alleged to have committed a serious act of misconduct where "there was an escalating pattern of less serious misconduct, which could have been abated through intervention." The Standard suggests but does not require the use of such performance indicators as use-of- force incidents, citizen complaints, and disciplinary actions. • A 1989 report by the International Association of Chiefs of Police (IACP) (1989:80) recommends EI systems as a means of controlling corruption and building integrity in police departments.14 • The U.S. Civil Rights Commission was the first agency to recommend EI systems, in its 1981 report, Who is Guarding the Guardians?15 • Since 1997, EI systems have been included in consent decrees and memoranda of understandings settling law suits brought by the Civil Rights Division of the U.S. Justice Department under the "pattern or practice" clause of the 1994 Violent Crime Control Act. EI systems are mandated in the agreements related to the Pittsburgh Police Bureau; the New Jersey State Police; the Metropolitan Police Department of Washington, D.C.; the Los Angeles Police Department; and the Cincinnati Police Department.16 The consent decree settling the Justice Department suit against the New Jersey State Police, for example, requires the development of a Management Awareness Program (MAP) for the purpose of "maintaining and retrieving information necessary for the supervision and management of the State Police to promote professionalism and civil rights integrity, to identify and modify potentially problematic behavior, and to promote best practices..."17 The consent decree with the Los Angeles Police Department (LAPD) sets requirements for an EI system, known as TEAMS (Training Evaluation and Management System). The consent decree specifies seventeen categories of data to be collected and entered into the system.18 BOX 1.2 EI Systems Recommended By: • U.S. Department of Justice • U. S. Civil Rights Commission • Commission on Accreditation for Law Enforcement Agencies • International Association of Chiefs of Police EI Systems and Other Best Practices EI Systems are the centerpiece of an emerging package of best practices designed to enhance police accountability. The other elements in the package of best practices, as recommended by the U.S. Department of Justice report Principles for Promoting Police Accountability (2001), include a comprehensive use-of-force reporting system and an open and accessible citizen complaint system. Data on officer traffic enforcement can also be entered into an EI system and can be used to address the issue of potential racial bias in traffic stops.19 The EI system is the central repository of data on use-of-force and citizen complaints (and other indicators). In this new paradigm of accountability, particular issues such as use-of-force performance are no longer treated as separate items but are linked to all areas of officer performance in a systematic fashion. A Vera Institute evaluation of the implementation of a consent decree in Pittsburgh commented that "the early warning system is the centerpiece of the Police Bureau's reforms in response to the consent decree."20 It is the mechanism by which these performance data become useful to managers committed to enhancing accountability and reducing officer misconduct.21 Goals and Impacts of EI Systems The goals and potential impacts of EI systems are broader than generally understood. When the EI concept originally developed more than twenty years ago, EI systems had a narrow focus on "problem" police officers. The implicit idea was to "catch" the "bad" cops. As EI systems have developed over time, however, police managers increasingly understand that they have broader goals and potential impacts, including individual officers, supervisors, and the department as a whole. Individual Officers EI systems have a major focus on individual officers. The goal of the system is to correct the performance of officers who appear to be having performance problems. The NIJ study found that the EI systems investigated did succeed in reducing citizen complaints and officer use of force among officers subject to intervention. This project and the findings are discussed in detail in Chapter Two.22 One important change that has occurred is that there is now a broader definition of a "problem" employee. Originally, problem officers were defined primarily as those who frequently used excessive force and received a large number of citizen complaints. Reflecting a broader approach, however, the commander responsible for the Pittsburgh Police Bureau's PARS reports that the system allows them to identify their "top performers" and their "under-performers" as well as their "problem officers". Another police department with a comprehensive system, for example, identified an officer who had made no arrests, no pedestrian stops, and no traffic stops in a given time period. The data also indicated that this officer was working the maximum number of permissible hours of off-duty employment. These indicators allowed the department to take the appropriate corrective action. The capacity to identify top performers as well as officers with performance problems requires a comprehensive EI system such as the Pittsburgh PARS. Most of the early EI systems are not able to identify under-performers, however, because they do not collect the necessary performance data, such as arrest activity. The difference between comprehensive and more limited EI systems is discussed in detail in Chapter Two. Supervisors An EI system also reorients the role of supervisors in a department. An EI database provides first-line supervisors with documentary evidence of officers' performance and permits both a detailed analysis of individual officers and comparisons among officers. In the survey of police managers (See Chapter Four) one manager reported that the EI system as "A useful tool to involve supervisors and lieutenants in non-traditional models of problem solving. It has served to enhance their management skills and help round out their people interaction skills." At the same time, however, orienting supervisors to their new role is one of the major challenges facing EI systems. Engaging an officer with performance problems and helping that person to recognize his or her performance problems is a role that many supervisors are reluctant to undertake. Some prefer to define their role as friend and supporter of their officers. Some do not have the skills for the task. Most have not been properly trained for this new role. Supervisor training tends to emphasize the formal disciplinary aspects of the job. The intervention phase of an EI system requires supervisors to be a combination of coach and trainer, providing a delicate mix of support, criticism, and help.23 As already mentioned, the Pittsburgh PARS takes a special approach to reorienting the role of supervisors. Sergeants are required to access the system's database and review the performance of officers under their command. In addition, as mentioned earlier, the PARS requires sergeants to conduct "roll-bys" of officers who have been identified by the system. In this way, PARS creates a structured and systematic form of intensive supervision for officers who have been identified as having performance problems. (In this and other systems, a supervisor has a password granting access to all officers under his or her command but no other officers. A district commander at the rank of captain, for example, would have access to the records of all lieutenants, sergeants, and officers in that district.) Some EI systems record each supervisor's access of the system. Thus, the performance review of a sergeant can include a review of the extent to which he or she accessed the system during the period under review. An evaluation of the consent decree in Pittsburgh concluded that the department's PARS "marks a sweeping change in the duties of the lowest level supervisors." Supervisors log on to the PARS each day prior to roll call. The screen notifies each supervisor of any critical incident involving an officer under his or her command since the last time that supervisor logged on.24 The consent decree regarding the Los Angeles Police Department requires that "on a regular basis, supervisors review and analyze all relevant information in TEAMS II about officers under their supervision...."25 Additionally, the consent decree requires that "LAPD managers on a regular basis review and analyze relevant information in TEAMS II about subordinate managers and supervisors in their command...and that managers' and supervisors' performance in implementing the provisions of the TEAMS II protocol shall be taken into account in their annual personnel performance evaluations."26 The consent decree involving the Cincinnati Police Department contains a similar requirement that supervisors be evaluated on the basis of "they're to use the risk management system to enhance department effectiveness and reduce risk"27 BOX 1.3 Goals and Impacts of EI Systems The Individual Officer • Improved Performance • Higher Standards of Accountability Supervisors • Improved Supervisory Practices • Higher Standards of Accountability as Supervisors The Department • Higher standards of accountability • Reduction in litigation • Improved Community Relations EI systems can also address the problem that, with regular shift changes, sergeants often find themselves responsible for officers who they do not know. As one commander reports, "There is a lot of movement of personnel, so supervisors often do not know the histories of their officers. The EWS report brings them up to speed in a much more timely fashion." Chapter Four of this report contains the findings of a survey of police managers with EI. The San Jose Police Department has taken EI systems to a new level by developing a Supervisor's Intervention Program (SIP) addresses the performance of supervisors as well as rank and file officers. Whenever the team of officers under a supervisor's command receives three or more citizen complaints within a six- month period, the supervisor is required to meet with his or her chain of command and the head of Internal Affairs. (The San Jose EI system relies only on citizen complaints.) In the first year of the SIP program, four supervisors met the thresholds and were counseled by the department.28 Some departments are exploring ways to use their EI system to hold sergeants accountable. The San Jose Police Department, for example, identifies sergeants when officers under their command receive a certain number of citizen complaints. This approach recognizes that many officer performance problems are the result of inadequate supervision.29 The Department An EI system also has the potential for considerable impact on the department as a whole. The system defines standards of conduct and provides a database for measuring officer performance and identifying substandard performance. At the same time, a fully operational system redefines the role of supervisors, giving them specific duties related to supervision of officers with performance problems. And as mentioned in the previous section, the more sophisticated EI systems have the capacity to monitor supervisors' use of the system and to grade their performance accordingly. Problems identified by the EI system can also lead to changes in departmental policy. The consent decree between the Justice Department and the New Jersey State Police, for example, requires that "Each supervisor shall, consistent with his or her authority, implement any appropriate changes or remedial measures regarding traffic enforcement criteria, training, and enforcement practices for particular units or subunits..."30 Over the long run, the cumulative effect of the changes in the role of supervisors and changes in departmental policy has the potential for changing the organizational culture of a police department and establishing new standards of accountability. In the police commander survey (See Chapter Four), one commander explained that over time rank and file officers came to accept the goals of the EI system: "Most no longer believe this system is out to 'get' them, but rather to assist them." A top commander in the Pittsburgh Police Bureau claims that PARS has begun to change the culture within the department. Two impacts are particularly notable. First, the system forces a more intensive form of supervision on the part of sergeants. Second, by identifying officers who are underperforming because they are devoting too much time to off-duty employment, and then correcting that problem, PARS helps to develop a greater commitment by officers to their primary employment.31 The positive impact on the organizational culture of a department is not guaranteed, however. If there is little effective use of the EI system–as in failure to conduct meaningful interventions with officers–the impact could easily be negative: reinforcing officer cynicism about the gap between stated ideals and actual practice Learning from the Corrections Profession The corrections profession has already begun to incorporate the EI concept into its accreditation process. The American Correctional Association (ACA) is adopting a "performance-based" approach to accreditation. The indicators for agency performance are essentially the same as those used in law enforcement EI systems. The Performance-Based Standards for Adult Community Residential Services, for example, includes the following standard:32 "Number of offenders' grievances filed alleging inappropriate use of force in the past 12 months, divided by [the] Average daily offender population in the past 12 months.” The approach used by the ACA focuses on the agency, with organizational-level performance measures, while law enforcement EI systems focus on individual officers. There are good reasons for suggesting that the ACA approach is the better of the two. It properly focuses on organizational-level issues of management and supervision where corrective action can prevent future misconduct. The current law enforcement focus on individual officers tends to make scapegoats of a few officers, generate hostility from the rank and file, and ignore important issues of management and supervision. Early Intervention and Community Policing EI systems are fully compatible with the goals of community oriented policing (COP). In particular, the community policing movement has created demands for new measures of police performance.33 Because COP reorients the goals of policing, new measures of are needed for both individual police officers and entire departments. The traditional measures of the crime rate and the clearance rate are no longer adequate. Traditional police performance measures are inadequate in several regards. They emphasize crime, to the neglect of quality-of-life issues; they fail to take into account perceived community needs; and they fail to reward adequately good police performance.34 By systematically identifying and attempting to control inappropriate behavior, an EI system can potentially reduce the number of incidents that alienate communities from the police. One of the basic goals of COP is to establish closer ties to the communities receiving police services. COP places particular emphasis on being more responsive to community concerns about the quality of police services, particularly with respect to racial and ethnic minority communities concern about excessive force or other forms of inappropriate behavior by officers. In this regard, Alpert and Moore recommend "the development of statistical evidence on the use of force and the incidence of brutality, discourtesy, and corruption..." This is precisely what an EI system does. Indeed, Alpert and Moore's approach suggests that a law enforcement agency's personnel data system should transcend a narrow focus on suspected officers with performance problems and include all current sworn officers.35 Early Intervention and Problem-Oriented Policing EI systems are also consistent with the principles of problem- oriented policing (POP). In this case, the problem is not graffiti or public drunkenness, but those officers with performance problems. As initially formulated by Herman Goldstein, POP holds that law enforcement agencies should desegregate the various aspects of their role and, instead of attempting to address "crime" and "disorder" as global categories, should identify particular problems within each category and develop narrowly tailored responses appropriate to each. In addition, they should develop the appropriate performance measures for each problem.36 The POP process of scanning, analysis, response, and assessment (SARA) is, for all practical purposes, the way an EI system operates: • Scanning: review of data in the EI system database. • Analysis: identification and selection of potential officers with performance problems. • Response: intervention with selected officers. • Assessment: post-intervention review of officers' performance. Early Intervention and COMPSTAT EI systems are very similar to COMPSTAT programs in terms of both purpose and process. Both rely on the analysis of systematic data for the purpose of addressing problems and holding police managers accountable for problems under their command. An evaluation of the Pittsburgh PARS (its EI system) described one key element as modeled after COMPSTAT. The Quarterly COMPSTAT meetings in Pittsburgh involve a review of officers who have been identified by PARS. Area commanders make presentations about any officers under their command who have been identified and conclude with a recommendation regarding formal intervention. After a discussion, the chief of police makes a final decision on what course of action to take.37 COMPSTAT is one of the most notable police innovations in recent years, and has been widely adopted across the country. A COMPSTAT program involves a computerized database of timely and systematic data on criminal activity. Analysis of the data provides a means of developing rapid and narrowly targeted responses designed to reduce crime. Regular COMPSTAT meetings are designed to hold police managers accountable by requiring them to describe crime trends in their areas of responsibility and to explain what responses they have developed. Early Intervention and Risk Management EI systems can be a key component in a risk management system (RMS).38 Risk management is a process for reducing an organization's exposure to financial loss due to litigation. In law enforcement, the major civil litigation costs arise from fatal shootings, excessive physical force, and high-speed pursuit incidents. An EI system database can readily identify patterns of individual officers and situations that represent actual or potential risks. It also provides a structured process for correcting those problems and reducing the potential financial risk. In fact, the expanded EI system in the Cincinnati Police Department, as mandated by a memorandum of understanding with the Justice Department, is called the Risk Management System. Early Intervention and Police-Community Relations EI systems represent a potentially effective response to the historic problems of officer misconduct and tensions between law enforcement professionals and the communities they serve. For many decades, alleged police officer misconduct, including misuse of deadly force, use of excessive physical force, and discourtesy, has been a major cause of tensions between the police and racial and ethnic minority communities. Civil rights leaders have alleged that minority citizens are not only the targets of police misconduct at a rate disproportionate to their presence in the population, but that police departments have failed to investigate citizen complaints about misconduct and discipline guilty officers.39 To an unfortunate extent, police abuse of citizens has been explained in global terms, typically labeling all police officers with a broad brush. Abusive behavior for example, has been attributed to a general police subculture, with the implicit assumption that certain attitudes and behaviors are common to all officers in all police departments.40 Other observers have attributed abusive behavior to race, arguing that abuse reflects racist attitudes and behavior on the part of white police officers toward citizens of color.41 Some observers have attributed overly aggressive or abusive behavior to gender, holding that reflects male norms of behavior.42 Still other observers have attributed police misconduct to organizational dysfunction, arguing that poor leadership and low standards of professionalism have tolerated many different forms of police officer misconduct.43 The great virtue of an EI system is that it can pinpoint specific actions that are creating conflict with the community and the particular officers responsible for them. For example, if there are complaints in one neighborhood about frequent use of force, an EI system can help to confirm or refute the allegations, and if confirmed, identify the officers involved. At the same time, it can help to identify those officers who are engaged in active police work (e.g., a high volume of arrests) without resorting to inappropriate behavior. Early Intervention and Racial Profiling A comprehensive EI system has the potential for addressing the issue of racial profiling. The Pittsburgh and Phoenix EI systems collect data on traffic stops, including data on the race and gender of drivers. These data make it possible to identify officers who are stopping a suspiciously high number of racial or ethnic minority drivers. As a discussion paper by Samuel Walker argues, an EI system solves many of the problems associated with interpreting traffic stop data. Rather than use resident population data, as many reports do, an EI system permits comparisons among officers working similar assignments.44 In a predominantly Latino neighborhood, for example, it is to be expected that a high percentage of all stops would involve Latino drivers. The EI system database can help identify officers who are stopping far more Latino drivers than their peers. The data alone would not prove that racial profiling exists. Rather, it would be the starting point for supervisory review that would determine whether or not an officer's activities involve racial or ethnic bias. The peer officer comparison approach is discussed in Chapter Two of this report. It is the methodology currently used in the Pittsburgh PARS.45 In one department, the EI system identified an officer who had a practice of stopping female drivers for questionable purposes. The department received a formal citizen complaint from one female driver who alleged that the officer made improper sexual advances during a traffic stop. A review of the EI database revealed a very high rate of traffic stops of female drivers. Combined with the original complaint, these data were sufficient to cause a formal departmental intervention. Using peer officers as the comparison group solves the most vexing problem with regard to the analysis of traffic stop data. Most current data collection efforts use the racial and ethnic composition of the resident population as the baseline or denominator. One Size Does Not Fit All Departments There is no single model for EI systems. The outline of the basic components of a system described above represents a single framework. Each department can develop an EI system appropriate to its needs. These needs will vary significantly according to the size of the department.46 Medium-sized and small agencies do not need as complex and sophisticated a system as do large departments. Most of the available information about EI systems at present, however, is derived from large law enforcement agencies. Consequently, at this time it is not possible to specify the exact needs of medium- sized and small departments. Developing those needs is on the list or priorities for future research and program development (See Chapter Five). A Work in Progress Although EI systems are increasingly popular and have been recommended by a wide range of professional groups, they are far more complex than is generally realized. They require making a number of difficult choices regarding their design, and once operational require close, ongoing administrative attention. EI systems are in a state of continuing development. There is no solid consensus regarding a single best way to operate a system. Local police departments have found that they need to continually revise and fine-tune their EI systems. As one police manager explains, "this is a new practice that may, and has, changed over time. It continues to evolve." Another explains that an EI system "needs to continuously evolve and expand" (See Chapter Four). Additionally, our understanding of EI systems is experiencing a paradigm shift. First, they are no longer seen only as a means of identifying officers with performance problems, but instead are recognized as an effective data-based tool for general personnel assessment. Second, instead of focusing exclusively on rank and file officers they are seen as a tool for promoting the accountability of supervisors. Conclusion Early Intervention systems are an important new tool for police accountability. They are data-based programs for identifying officers with performance problems and providing the basis for departmental intervention to correct those problems. EI systems have emerged as a recommended best practice in police accountability. They have been recommended by a wide range of professional associations. In addition, EI systems are consistent with the goals of both community policing and problem-oriented policing. Although they are growing in popularity, EI systems are still in the early stages of development. Departments are still wrestling with a number of difficult and important issues. These issues are addressed in the remaining chapters of this report. Chapter Two The Components of an EarlyIntervention System The Components.................................................25 Size and Scope: Two Approaches to EI Systems ..................25 Performance Indicators.........................................26 The Number of Indicators.......................................26 Department-Specific Indicators ................................29 Identification and Selection Process ..........................30 Identifying Problem Performance: Three Examples................30 The Proper Thresholds .........................................31 Mandatory vs. Discretionary Referral to Intervention ..........33 Feedback to Officers...........................................35 Intervention...................................................35 Action Alternatives............................................35 Training Supervisors for Intervention..........................37 Holding Supervisors Accountable ...............................38 Post-Intervention Monitoring...................................39 An EI System Appropriate for Each Department ..................40 The Need for Careful Planning..................................41 Conclusion ....................................................41 The Components of an Early Intervention System The Components An Early Intervention system consists of four basic components: performance indicators, identification and selection process, intervention, and post- intervention monitoring. An earlier report identified three components of EI systems.47 This report, reflecting the maturation of EI systems, utilizes four components, because the identification and selection process is a separate component from the list of performance indicators. Size and Scope: Two Approaches to EI Systems While all EI systems will have the same basic components, the size and scope of different systems will vary according to their goals.48 Two basic types of EI systems currently exist, representing differently sized systems with different goals. Some systems, such as the Pittsburgh PARS and the Phoenix PAS, are comprehensive personnel assessment systems. They collect a very wide range of data and are able to conduct complex data analyses that can include identifying top-performing and underperforming officers as well as officers with performance problems. Systems of this sort not only require a sophisticated technological infrastructure but also an enormous amount of administrative commitment.49 BOX 2.1 Components of an EI System • Performance Indicators • Identification and Selection • Intervention • Post-Intervention Monitoring Many other systems, such as the Miami-Dade and the Tampa, Florida systems, are more narrowly focused and may be described as performance problem systems. They collect a smaller range of data and focus on a more limited range of performance issues. As a result, they are also less expensive and require a smaller commitment of administrative resources. Neither system is necessarily better than the other. They simply have different goals and different approaches to police accountability. The goals of the more comprehensive systems are more ambitious, but they are also more demanding of the department. An evaluation of Pittsburgh's PARS, for example, pointed out that its comprehensive approach comes at a price. By imposing greater administrative responsibilities on sergeants for monitoring officer behavior, it necessarily takes time away from traditional field supervision and other duties.50 Performance Indicators The performance indicators involve those aspects of officer performance that are the subject of official departmental reports and are entered into an EI system database. Almost all experts argue that an EI system should utilize a broad range of performance indicators and not rely on just one indicator as many of the first EI systems did (and some still do). Relying on a single indicator such as citizen complaints is not recommended. There are a number of problems related to official data on citizen complaints and they are not consistently reliable indicators of officer performance.51 Using a broader range of indicators is more likely to identify officers whose behavior requires departmental intervention. The Number of Indicators There is no consensus of opinion regarding the exact number of performance indicators to use. The Pittsburgh PARS uses 18 performance indicators (although the department is considering adding five more). The Phoenix PAS utilizes 24 indicators. The consent decree between the U.S. Department of Justice and the Los Angeles Police Department specifies 17 categories of data to be entered into the department's TEAMS II (See Appendix A).52 The Memorandum of Understanding involving the Cincinnati Police Department requires the collection of ten performance indicators. Box 2.2 represents a list of performance indicators in a model comprehensive personnel assessment system. Box 2.3, meanwhile, lists the performance indicators in a model problem officer system. BOX 2.2 Performance Indicators Comprehensive Personnel Assessment Model 1. All non-lethal uses of force Citizen injuries requiring medical attention Disciplinary outcomes 2. All officer-involved shooting incidents On-duty/off-duty discharges, citizen injuries, citizen fatalities Disciplinary outcomes 3. All officer-involved vehicular pursuits Accidents Injuries Fatalities Disciplinary outcomes 4. All citizen-initiated complaints Use of force Racial bias Investigative outcomes 5. All citizen-initiated commendations or compliments 6. All departmental commendations and awards 7. Criminal arrests and investigations of subject officer Chapter 2 BOX 2.2 Performance Indicators (Continued from page 27) 8. Civil suits or an administrative claim in which subject officer is named Nature of claims Case outcomes 9. All arrest reports, crime reports, and citations made by officers 10. All motor vehicle stops and pedestrian stops made by subject officer Race, ethnicity, gender, age of drivers stopped 11. Performance evaluations for each officer 12. Training history Any failure of an officer to meet weapons qualification requirements 13. All management and supervisory actions, including non-disciplinary actions, related to each officer 14. Sick leave or family leave record 15. Canine unit deployment involving subject officer Deployments Bites Citizen injuries Complaints or civil suits arising from deployments 16. Failure to appear in court BOX 2.3 Early Intervention System Focused Performance Model 1. All officer-involved use-of-force incidents 2. All officer-involved shooting incidents 3. All citizen-initiated complaints 4. All officer-involved civil suits or administrative claims 5. All departmental disciplinary actions 6. All citizen-initiated or department-initiated commendations or awards. The choice of the number of indicators to use involves a trade-off. The greater the number of indicators used, the greater the capacity of the system to evaluate the full range of officer performance. The more comprehensive Pittsburgh PARS, for example, allows commanders to identify top-performers and under-performers, as well as officers with performance problems. Because the system includes arrest data, commanders can identify officers who have several use-of-force reports but who are also making a very large number of arrests and compare them with officers who have slightly fewer complaints but who make very few arrests. At the same time, however, utilizing a greater number of indicators imposes greater administrative demands in terms of data entry and data analysis.53 A smaller problem officer system, with fewer performance indicators, will be much less expensive and easier to create and maintain. Department-Specific Indicators Some performance indicators have been adopted because of specific controversies surrounding a particular department. The memorandum of understanding between the U.S. Department of Justice and the City of Cincinnati requires that the Risk Management System include drawing and pointing a weapon at a citizen. This indicator is included because there had been many complaints from citizens about such action by police officers.54 The Phoenix PAS includes High Stress Radio Calls (calls involving actual or intended suicides, or child abuse, molestation, or death) as a performance indicator. This indicator is being included because of three incidents involving suicide or attempted suicide by officers following a series of calls to incidents of this sort. In one case, an officer mentioned to his supervisor that he had seen enough of these calls, and expressed suicidal thoughts.55 Individual departments may have other special problems they are aware of and want to incorporate into their EI systems. Identification and Selection Process Given the existence of a performance database, a strategy is needed to identify officers for performance review and to select officers for referral to intervention. These are separate stages. In many EI systems an officer who is identified will automatically be referred to intervention. Others, however, utilize a two-stage process through which officers who are identified are subject to a full-scale performance review with the result that only some are referred to formal intervention. Several participants at the Early Intervention State of the Art Conference56 vigorously emphasized the point that a computerized database only gives you numbers. It does not make decisions about what to do with identified officers; only supervisors can do that. These statements only highlight the point that there is no magic in a sophisticated computerized system. To be effective, an EI system requires responsible management by the command staff. Identifying Performance Problems: Three Examples The following examples are based on actual cases adapted from the EI system of one large police department.57 Officer A The Indicators: Officer A had five use-of-force reports during one reporting period. Relative to other officers in the same unit, this was a very low figure. But, he made only eight arrests during this period. The Analysis: Officer A used force in more than half of the arrests he made. The ratio of force to arrests is cause for departmental concern and probable intervention. Officer B The Indicators: A female driver filed a citizen complaint against Officer B alleging an inappropriate sexual advance during a traffic stop. An examination of Officer B's EI file indicated a very high number of traffic stops involving females. The Analysis: Officer B appears to be abusing his law enforcement powers to harass female drivers. Officer C The Indicators: Officer C had no citizen complaints or use-of-force reports for the reporting period. He also had made no arrests, no traffic stops, and no field stops. An examination of his EI file indicated that he was working the maximum number of permitted hours of off-duty employment. The Analysis: Officer C. was devoting all his energy to his off- duty employment rather than fulfilling his responsibilities to the department. The Proper Thresholds The central issues with regard to identifying and selecting officers for intervention the question of the proper performance thresholds to be used. (Some systems use the term "triggers" but that word has inappropriate connotations and is not recommended.) The issue can be framed in the following terms: ** How many use-of-force incidents over what time period should qualify an officer for intervention? ** What combination of use-of-force reports, citizen complaints, and other indicators over what period of time should qualify an officer for intervention? At present, there is no consensus regarding the best set of thresholds. Three different formulas are currently being used: department-level thresholds, peer officer thresholds, and ratio- based formulas. 1. Department-Level Thresholds In the Miami-Dade EIS, an officer is identified if in any quarter he or she has two or more citizen complaints or three or more use-of-force reports in the previous 90 days (or quarter). A list of the officers so identified is circulated to commanders who then must take one of four actions: refer the officer to the Psychological Services Bureau for counseling, refer the officer to the department Stress Abatement Program, refer the officer to some other form of corrective action, or conclude and document that no action is required. The Miami-Dade EIS also operates at an earlier and less formal manner than the formal process described above. Monthly reports are circulated listing officers who have two or more citizen complaints or two or more use-of-force reports (or both). Commanders are not required to take any formal action, but are expected to review an officer's performance and perhaps conduct an informal meeting with him or her. 2. Peer Officer Averages The Pittsburgh PARS utilizes an approach that compares officers performances with peer officers, meaning officers with the same assignment (e.g., night shift, high-crime area, daytime traffic unit, etc.). Some experts have described this approach as comparing apples with apples. Officers assigned to high-crime areas are more likely to generate use-of-force reports and citizen complaints because of the nature of the work environment. It is unfair to expect them to meet the same numerical standard as officers working low crime areas.58 The Memorandum of Understanding mandating the EI system in the Cincinnati Police Department (officially called the Risk Management System) also requires a peer officer formula. The RMS is required to report data on the average level of activity for each data category by individual officer and by all officers in a unit (See Appendix D). Finally, the Phoenix PAS is developing a peer-officer-based formula. 3. Performance Indicator Ratios A third approach to thresholds analyzes performance data in terms of the ratios between different performance indicators: e.g., the ratio of use-of-force reports to arrests. In one example, (See Box 2.4, Officer A), an officer did not use force very often, but used force in more than half of all arrest situations. One of the virtues of this approach is that it allows commanders to identify top- performing officers (e.g., high arrest activity, low arrest-to-complaint ratio) and not unfairly subject them to intervention. The Phoenix PAS has been testing this approach prior to implementation.59 Mandatory vs. Discretionary Referral to Intervention Some systems use a mandatory approach for referral to intervention while others use a two-stage discretionary approach. In the initial Minneapolis Police Department Early Warning System, an officer was automatically referred to intervention if he or she received three complaints in a twelve-month period.60 Not only did this system (which has since been replaced by a more sophisticated performance review system) rely only on citizen complaints but it did not take into account the nature and context of the complaints. The Miami-Dade EIS generates monthly reports that list officers who have received two or more citizen complaints or have made two or more use-of-force reports (or both) in the last 60 days. This list is for informational purposes only, and supervisors are not required to take any action. It is understood, however, that supervisors will review any listed officer's performance and possibly speak to individual officers. Quarterly reports list officers who have two or more personnel (e.g., citizen) complaints and involvement in three or more use of force incidents (or both) within the previous 90 days. Supervisors must take action with regard to these officers. The alternatives include referring the officer to the stress abatement program or to the Psychological Services Program for counseling, or requiring some other corrective action, or concluding (and documenting) that no action is required. In short, commanders are provided information about officer performance but are left the discretion to decide whether any formal intervention is required, and if so what action is appropriate. The Los Angeles Sheriff's Department's PPI system utilizes a discretionary two-stage approach. Only some of those officers initially identified by the system are placed on Performance Review. Between 1996 and 2002, a total of 1,213 employees were identified, but only 235 (or 19 percent) were placed on Performance Review.61 The Performance Review Committee screens officers twice. First, the committee solicits the views of an officer's captain about whether placement on Performance Review is appropriate. A second screening, taking into account the captains' views, is conducted to select officers for a detailed Employee Profile Report (EPR) (known informally as a "Blue Book").62 The EPRs are prepared by sergeants assigned to the department's Risk Management Bureau. The performance review is designed to place an officer's performance in the context of his or her assignment and performance history. It considers evidence that might exculpate an officer along with evidence indicating a problem that is believed to be worse than suggested by the performance data. For example, an officer with a number of use-of-force reports might be excused because of particularly intensive law enforcement activity during the period under review and an otherwise exemplary record. At the same time, another officer might have in addition to a number of complaints and use-of-force reports a reputation for a bad temper and inability to work well with colleagues. In the end, some of the identified officers are selected for intervention. Pittsburgh's PARS also generates information about officer performance for the benefit of supervisors but leaves them the discretion to decide what action to take. An evaluation of the PARS found that formal intervention is a rare outcome for officers initially identified by the system. The most common response involves monitoring of the officer through supervisory field checks (also referred to as "roll-bys").63 Part of the reason for the few formal interventions is that, unlike other EI systems, the PARS has a formal process for pushing responsibility for monitoring officer behavior down to first line supervisors. As explained elsewhere in this chapter, the PARS requires sergeants to log on to the system every day, a process that forces daily consideration of performance data. Feedback to Officers One issue that has been discussed among experienced EI system commanders is the nature of feedback to officers who are identified but not selected. There may be some value in informing such officers that their performance was identified by the EI system but that upon review, it was determined that there was no problem requiring formal intervention. This situation is likely to arise with high-performing/low-ratio officers. Informing them of the decision not to refer them to intervention may be a valuable way of communicating departmental approval and support for their performance. Intervention The intervention phase consists of some form of counseling or retraining for officers who have been selected. In some EI systems the counseling is done by an officer's immediate supervisor. In other departments, the counseling involves other command officers. Action Alternatives The Commission on Accreditation for Law Enforcement Agencies (CALEA) Standards state that a departmental Personnel Early Warning System should have a menu of remedial actions for officers selected for intervention.64 In the Miami-Dade Police Department, the action alternatives include referral to the departmental Psychological Services Program or participation in the Stress Abatement Program. Another alternative is a determination that no problem exists and that no formal action is necessary. One of the virtues of the data-driven nature of EI systems is that the performance data provide the basis for discussing specific aspects of an officer's performance. Chapter Four of this report includes comments by police managers explaining how their EI systems strengthen the hand of supervisors in dealing with officers with performance problems. Where citizen complaints about rudeness are frequent, referral to Verbal Judo might be the proper response. Where problems in handling arrest situations are the problem, specific retraining can be offered. Depending upon a department's existing personnel procedures, the counseling or retraining can be recommended or mandated. In the Los Angeles Sheriff's Departments PPI system, the Performance Review Committee develops a formal performance plan for each subject officer.65 Intervention Group Training Classes Some EI systems conduct intervention through classes for groups of selected officers.66 There are a number of problems with this approach, however. First, the subjects covered in a class are general in nature and do not necessarily address the particular problems of individual officers. Second, they are difficult to schedule and considerable time may pass between the time an officer's problems are identified and he or she is enrolled in the intervention class. Third, there is evidence that bringing a group of selected officers together creates a dynamic of group solidarity that reinforces inappropriate attitudes. Such classes have been referred to as "bad boys" classes.67 Training Supervisors for Intervention The most important issue identified at the Early Intervention State of the Art Conference involved preparing supervisors to conduct interventions.68 Lack of adequate orientation and training was also one of the most serious problems reported in the police managers survey (See Chapter Four). One explained that as a result of inadequate training, the system is misunderstood by the first line supervisors, and therefore retraining is needed at the application level. EI systems represent a new role for supervisors, especially sergeants, in which they are expected to coach and help officers under their command. The traditional role of sergeants has involved formal disciplinary actions, and training has emphasized the procedural aspects of the process. In an EI system it is essential that first-line supervisors actually conduct intervention sessions, and do so in a manner consistent with the system's goals. It is possible for a particular sergeant to simply tell an officer not to worry about it. This would undermine the EI system and breed cynicism among officers. A related problem is the potential for favoritism, in which one supervisor excuses the performance of one officer under his or her command while taking a tough stand on another officer with similar performance problems. A final problem is the possibility of inconsistent interventions across a group of supervisors. Since interventions are designed to be confidential, it is not possible to require documentation of the actual substance of intervention sessions. It is essential that departments provide proper training for supervisors regarding their role in interventions and that this be done well before an EI system become operational. Several steps are possible to ensure that interventions are delivered as desired. First, in some departments the intervention is conducted in a meeting with more than one supervisor. This helps to ensure that one supervisor will not inappropriately excuse the performance of an officer. At least one system in operation is programmed to require supervisors to notify their own supervisors that the intervention has occurred. Failure to do so will result in an automatic email notice to the higher-ranking supervisor. This procedure, of course, does not guarantee anything about the content of the session. BOX 2.4 Intervention Action Alternatives • Referral to department employee assistance program • Referral to professional family counseling service • Referral to credit counseling services • Retraining on traffic stop tactics • Verbal judo training Holding Supervisors Accountable Some EI systems have begun to address the issue of inadequate first-line supervisors. If one or more officers in a unit are identified by an EI system, part of the problem may be inadequate supervision. The San Jose Police Department has developed a Supervisors Intervention Program (SIP). A supervisor is identified by the system whenever officers under his or her command receive a combined total of three or more complaints in any six-month period. The supervisor is then required to meet with his or her immediate supervisor, the head of the Professional Standards and Conduct Unit (PSCO), and an assistant chief. One limitation of the system is that it relies on only one performance indicator, citizen complaints.69 The San Jose Police Department is perhaps the first to institute a formal process for using the EI system to evaluate the performance of supervisors. More discussion and program development on this issue is warranted in the immediate future. The Pittsburgh PARS takes a different approach, integrating the EI system into routine supervision. Sergeants are required to access the PARS database every day and to review the performance of all officers under their command. This approach socializes them into use of the system and the habit of data-driven supervision. Sergeants are able to intervene with an officer even before that officer crosses the system thresholds. Through this process, supervisors are expected to regard performance analysis and intervention as regular parts of their jobs and not some special and onerous task reserved only for serious problems.70 The Pittsburgh PARS also attempts to ensure that interventions are conducted through a process of computerized notifications. When an officer is identified by the system, his or her sergeant is notified by email that an informal counseling session is required. The sergeant must then report by reply email when the counseling session has been completed. Failure to provide this notification automatically generates an email notice to that sergeant's lieutenant. Failure of the lieutenant to take the proper action generates an email notice to the next higher commander. Post-Intervention Monitoring Following an official intervention, supervisors need to follow-up by monitoring the performance of an officer for a specified period of time. At least three different approaches to follow-up exist in current EI systems. These approaches vary by their degree of formality and record-keeping. In many departments, the monitoring is informal and conducted by immediate supervisors. Some departments keep officers on a list for a specified time period following intervention and continue to monitor their performance. In the Pittsburgh PARS, sergeants are required to do "roll-bys" of officers who have been identified by the system. A roll-by is a direct observation of an officer at work. In effect, this is a form of intensive supervision. An EI system such as Pittsburgh's has the effect of structuring and systematizing such intensive supervision. In the Los Angeles Sheriff's Department, an officer remains on Performance Review for two years after initial identification. An officer may be removed only upon recommendation of his or her captain and approval by the Performance Review Committee. In other departments, however, the monitoring involves an elaborate system of observation, evaluation, and reporting. The New Orleans PPEP involves a six-month follow-up in which supervisors are required to observe subject officers while they are on duty and to file a signed evaluation of officers' performance every two weeks. The resulting documentation ensures that monitoring occurs.71 The post-intervention follow-up component of EI systems has received the least attention by officials involved in developing and managing EI systems. It is an issue that needs additional attention in the immediate future. In determining the exact nature of the post-intervention follow-up, agencies also encounter the problem of balancing the demands of accountability with the need for efficiency. As we indicated in our discussion of this issue related to the performance database, there is no clear answer to the problem at present. It needs to be the subject of continued discussion and experimentation within the law enforcement profession. An EI System Appropriate for Each Department The range of requirements EI can address prohibits the development of any single definitive model of an EI system. The abovementioned basic system components can provide only a general framework. The most effective EI systems are developed specifically to address the needs of the personnel they serve, and those needs are as unique as the agencies that produce them. One of the most influential variables in EI system development is the size of the agency it will serve.72 Small- to mid-sized agencies do not require the same levels of complexity and sophistication as their larger peers. Large agencies not only have much different problem sets, however, they also have significantly different levels of resources available for EI system development. Due in large part to this discrepancy in resources, the vast majority of information currently available about EI system development and implementation is derived from work with larger law enforcement agencies. This makes for an unfortunate scarcity of information regarding the requirements for effective EI systems in smaller and mid-sized agencies. Developing this information is high on the list for future research and program development (See Chapter Six). The Need for Careful Planning It is the consensus of opinions among police commanders with extensive EI system experience that careful planning is necessary for the effective development and implementation of an EI system. One of the main points of this report is that EI systems are extremely complex administrative tools, with several critical components and several different goals. Several of the established EI systems have gone through significant revision since they were first implemented. The Phoenix Police Department took over two years to develop and implement its PAS. Other experts believe this was a wise decision. One important aspect of an effective planning process is to involve officers from all ranks in the process. This helps to ensure that officers fully understand and are committed to the goals and procedures of the EI system. Conclusion EI systems are extremely complex administrative tools. They consist of four basic components, each of which requires decisions about the exact design. Careful planning and consideration of these decisions is necessary for an EI system to be implemented effectively. A planning process is described in Chapter Five. Chapter Three Background: The Origins and Development of Early Intervention Systems Evidence About Officers With Chronic Performance Problems .............45 The Early History of EI Systems........................................48 The First Permanent Systems: Miami and Miami Dade .....................50 An Evolving Concept ...................................................53 An Emerging Best Practice..............................................54 Implementing EI Systems: A Major Challenge ............................55 The Los Angeles Sheriff’s Department PPI System........................55 The Los Angeles Police Department......................................57 Pittsburgh Police Department...........................................58 Research on the Effectiveness of EI Systems ...........................59 (1) The NIJ Study of Early Warning Systems.............................60 (2) Evaluation of Pittsburgh PARS .....................................67 (3) Assessment of the LASD PPI System..................................68 (4) Anecdotal Evidence From Other Departments .........................69 (5) The Police Managers’ Survey........................................69 Conclusion ............................................................70 Background: The Origins and Development of Early Intervention Systems The purpose of this chapter is to describe the origins and development of EI systems. The first section covers the basic rationale for EI systems: the argument that a small number of police officers are responsible for a disproportionate number of problematic incidents such as citizen complaints, use-of-force incidents, and civil suits against law enforcement agencies. The second section provides a brief overview of the history of EI systems. The final section examines the development of EI systems in a few selected jurisdictions for the purpose of illustrating the administrative difficulties some of them have had in implementing their EI systems. The final section summarizes the findings of a National Institute of Justice-sponsored evaluation of the EW systems in three major police departments. Evidence About Officers With Chronic Performance Problems Early Intervention (EI) systems originated out of a growing recognition that a few officers are responsible for a disproportionate number of any police department's problems related to use-of-force and citizen complaints. The existence of the so-called "problem officer" was known informally for a long time. The evidence was largely anecdotal, however, and departments did not take any official action to address the problem. Police chiefs increasingly commented that "90 percent of our problems are caused by ten percent of our officers."73 Professor Herman Goldstein, one of the leading experts on policing and the founder of problem-oriented policing, was the first expert to discuss the possibility of "Identifying Officers with a Propensity for Wrongdoing," and then doing something to improve their performance.74 He cited an experimental (but short- lived) program by Hans Toch in the 1970s in which peer officers counseled Oakland, California, police officers with records of use-of-force incidents.75 A 1981 report by the U.S. Civil Rights Commission, Who is Guarding the Guardians?, provided the first data documenting the phenomenon of a few officers receiving a disproportionate number of all citizen complaints. It published a table indicating that officers who had received five or more complaints over a two-year period represented only 12 percent of all officers receiving complaints but accounted for 41 percent of all complaints (Table 3-1). One officer had 12 complaints while two others had 11 complaints during this two-year period. In response, the Commission recommended that "A system should be devised in each department to assist officials in early identification of violence-prone officers."76 BOX 3.1 Houston Complaint Data Number of Complaints per Officer • 2 complaints 298 officers • 3 complaints 136 officers • 4 complaints 70 officers • 5 complaints 33 officers • 6 complaints 18 officers • 7 complaints 8 officers • 8 complaints 6 officers • 9 complaints 1 officers • 10 complaints 0 officers • 11 complaints 2 officers • 12 complaints 1 officers Source: U.S. Commission on Civil Rights, Who is Guarding the Guardians? A Report on Police Practices (Washington, DC: Government Printing Office, 1981), p. 166. The aftermath of the 1991 Rodney King beating heightened national awareness of the phenomenon of the problem officer. The Christopher Commission, appointed to investigate the Los Angeles Police Department following the King incident, identified 44 problem officers in the Los Angeles Police Department (LAPD) with extremely high rates of complaints.77 In a particularly notable observation, the Christopher Commission pointed out that these 44 officers were "readily identifiable" on the basis of existing departmental records. Yet, the LAPD appeared to have made no effective use of these records to identify the officers and take any corrective action. The Commission found, for example, that citizen complaint data were not used in making routine performance appraisals.78 A year after the Christopher Commission report, the Kolts Commission investigated the Los Angeles County Sheriff's Department (LASD), and using a similar methodology identified a group of 62 deputies with significant performance problems in the department. These 62 deputies were responsible for almost 500 use-of-force/harassment complaint investigations. Seventeen of these deputies were responsible for 22 civil lawsuits that resulted in damage awards or settlements of about $3.2 million against the County. The Kolts Commission concluded that the LASD had "failed to deal with officers who have readily identifiable patterns of excessive force incidents on their records." Not only were nearly all of the 62 deputies still on patrol duty at the time of the Kolts report, but–even more alarmingly–many served as field training officers (FTOs). The Kolts Commission warned that in that capacity they were "imparting their 'street wisdom' to patrol deputies."79 The report recommended the creation of an EI system within the LASD to identify these deputies. The Kolts report staff "were struck that the LASD conducted no regular or meaningful analysis of lawsuits and administrative investigations" that might help to avert further misconduct and save the county millions of dollars. It concluded that "the concept of 'risk management' was largely unknown and unpracticed in the LASD."80 Meanwhile, the Christopher Commission report spurred journalistic investigations of the problem officer phenomenon in other cities. The New York Times reported that in Kansas City, two percent of the sworn officers were responsible for 50 percent of all citizen complaints.81 A Boston Globe investigation found that 11 percent of Boston Police Department officers were responsible for 61.5 percent of all complaints.82 And in Washington, D.C., investigative journalists identified certain officers with long records of complaints and other problematic behavior.83 The first formal evaluation of EI systems confirmed the anecdotal evidence regarding problem officers. Sponsored by the National Institute of Justice (NIJ) (and directed by the author of this report), this study evaluated the EI systems in the Miami-Dade, Minneapolis, and New Orleans police departments. It found that officers who had been selected for EI system intervention had substantially more serious performance records than did peer officers in their original recruit classes. Officers selected by the Miami-Dade Employee Information System (EIS), for example, averaged twice as many use-of-force incidents per year as their peers and were twice as likely to have been reprimanded and three times as likely to have been suspended.84 A full discussion of the findings of this study appears at the end of this chapter. The Early History of EI Systems Initial Experiments EI systems began to emerge in the 1970s as a crisis management response to public concern over police abuse of force. The histories of most of these programs are lost and may not be recoverable.85 In its 1981 report, the U.S. Civil Rights Commission cited with favor the EI systems in Oakland, New York City, and Kansas City.86 Little is known about those systems except for the fact that none appear to have survived very long. The Mollen Commission report on corruption and violence in the New York City Police Department in 1984, for example, offers no evidence of a functioning EI system.87 Because of a Police Foundation evaluation, information about the original Kansas City program is available. The initial Kansas City early warning system was authorized in early 1972 and began operating in November, 1973, as part of an experimental Peer Review Panel Program.88 The Panel was part of the same effort to stimulate innovation in the Kansas City Police Department that included the well-known Kansas City Preventive Patrol Experiment.89 The Panel had the general goal of reducing tensions between the KCPD and the public and, to that end, sought to effect change in police officer attitudes and behavior through peer support and pressure. The program was designed to "identify those patrol officers with high frequencies of negative encounters with citizens." It used "no single fixed formula" for identifying officers with performance problems but utilized a combination of four separate criteria. These included: (1) three or more "negative encounters" with citizens, defined as both official citizen complaints and "interfering with an officer" charges; (2) referral by an immediate supervisor or other high-ranking department official; (3) an officer's "voluntary request for appearance;" (4) involvement in an incident of "public or departmental notoriety."90 The Police Foundation evaluation found that the Peer Review Panels had no significant effect on reducing either citizen complaints or interfering with officer charges among those officers who were subject to the program. The ultimate fate of the Kansas City Peer Review Panel is not known. Given the fact that it was an experimental program, funded through a grant from the Police Foundation ($73,000 in 1973 dollars), and had no positive impact, it appears that the program lapsed.91 A 1991 New York Times article on the Kansas City EWS portrayed it as a new program, with no reference to the one cited by the Civil Rights Commission.92 In Oakland, an experiment in utilizing police officers as change agents within their own department included one component designed to reduce citizen-officer conflict among officers with performance histories of frequent conflict. These officers were counseled by peer officers with exemplary performance records. The experiment concluded with some modestly optimistic conclusions about the effectiveness of peer counseling to deal with officers with performance problems.93 It appears, however, that once the formal experiment ended, the peer-counseling program ended as well. The 1981 report of the U. S. Commission on Civil Rights, Who is Guarding the Guardians?,94 which was based on extensive hearings on police misconduct in three major cities, took note of these early EW systems and recommended that all departments establish similar systems. Specifically, the Commission concluded that: The careful maintenance of records based on written complaints is essential to indicate officers who are frequently the subject of complaints or who demonstrate identifiable patterns of inappropriate behavior. Some jurisdictions have "early warning" information systems for monitoring officers' involvement in violent confrontations.95 The First Permanent Systems: Miami and Miami- Dade Two early EW systems that have had continuous operation from their creation to the present day, and about which detailed information exists, were developed in the City of Miami Police Department and the Miami-Dade (formerly Metro-Dade) Police Department in the late 1970s and early 1980s. The histories of these two systems reflect the process of development that has occurred in most of the early systems: the intersection of growing public pressure to curb misconduct and improve police- community relations and creative leadership by a few key individuals inside the department, operating without the aid of any external models. City of Miami In 1979, the City of Miami Police Department became concerned with its officers' behavior that generated citizen complaints in the late 1970s. In a May 29, 1979 memorandum to the chief, the commander of the Internal Security Unit suggested an early warning system based on concepts of organizational development. This involved a "cyclical model where the problem is diagnosed, external professionals are consulted, strategies are developed, programs are implemented and evaluated, and results are fed back to begin the cycle again."96 To illustrate his idea, the commander identified a list of officers, by assignment, who had two or more citizen complaints during a two-year period (1976-1978). He also compiled a list of officers who had received five or more civilian complaints during that period. Armed with these data and the Internal Security Monthly Activity Reports, the commander computed statistics indicating that five percent of all officers on the force accounted for 25 percent of all citizen complaints. He noted that "if this group were suddenly removed from our department, our complaint picture could be reduced by as much as one-fourth. Obviously, this group should warrant some special attention, if we are to reduce our complaint incidence."97 The data also indicated that the small groups of officers with the most complaints were also more likely to receive complaints regarding excessive force. The result of this analysis was a proposal that the Miami Police Department should systematically provide supervisors with information "that can be used to identify problem officers."98 The IA commander also noted that off-duty employment, including rock concerts, wrestling matches, and football games generated a high number of citizen complaints. He reasoned that fatigue may "heighten an officer's opportunity to react in an aggressive manner."99 He suggested that the department should respond to these officers before they become involved in self-destructive activities or develop a trend of violating departmental orders. His proposal included more intensive supervision, counseling by outside professionals and training in tactics and strategies:100 The Miami EW system eventually developed selection criteria including four categories of behavior: (1) citizen complaints, (2) control of persons (i.e., use-of-force) incidents, (3) official reprimands, and (4) firearms discharges.101 Once an officer is identified by the system, his or her supervisor is notified and is expected to meet with the officer and to determine if he or she needs any assistance, such as counseling, training, or other intervention. Officers identified by the EW system are then closely monitored. Supervisors are expected to investigate any problem incident and follow-up with a memorandum containing a recommendation. Internal Affairs provides the supervisor with a report of each incident that must be reviewed as well as the officer's assignment when the incident occurred. After evaluating the reports, the supervisor makes a recommendation that may include 1. Reassignment, 2. Retraining, 3. Transfer, 4. Referral to an Employee Assistance Program, 5. Fitness for Duty Evaluation, or 6. Dismissal Pursuant to Civil Rules and Regulations. The memorandum goes to the Commander of Internal Affairs through the chain of command. Each reviewing supervisor must agree or disagree with the recommendation.102 The success of the Miami EW system, however, is uncertain. In 2001-2002, the department faced a series of problems, including the indictment of a number of police officers on federal charges of abuse of authority. News media accounts indicated that some of the officers indicted had records of serious performance problems –which presumably should have been identified by the EW system. In late 2002, the chief of police resigned in the wake of these events.103 Miami-Dade The EW system in the Miami-Dade police department was developed at about the same time as the City of Miami's response to a number of racial incidents. (It is not clear what role the City of Miami's EW system played in influencing the development of the Miami-Dade system.) The beating of an African-American school teacher and the highly publicized fatal beating of Arthur McDuffie, an African-American insurance agent, by Miami-Dade officers, heightened racial tensions in the Miami area. On May 17, 1980, the acquittal of four officers accused of McDuffie's death provoked three days of riots that resulted in civilian deaths and millions of dollars in property damage.104 As a result of these problems, a local ordinance directed the Miami-Dade Police Department to develop an Employee Profile System (EPS) that provides detailed information on each employee. In addition to demographic data, the EPS includes all complaints, use-of-force incidents, commendations, discipline, and disposition of all internal investigations. As an adjunct to the EPS, the department implemented an Early Identification System (EIS) under the supervision of the Internal Review Bureau. The EIS was operational by 1982. Officers are identified on the basis of two or more "personnel complaints" and/or three use-of-force incidents in the previous three months, and then listed on the next Quarterly Report. The reports are disseminated to supervisors to be used as they deem appropriate with regard to each officer. The "Action Alternatives" include referral to counseling for stress, referral to a 40-hour training assistance program, or counseling by supervisory personnel. The EIS has remained basically unchanged for the past two decades.105 The NIJ evaluation of the Miami-Dade EIS found that intervention succeeded in substantially reducing use-of-force among subject officers.106 An Evolving Concept Without any guidance from recommended models, EI systems developed in an ad-hoc fashion. Over the years, a rough consensus began to emerge about the major system components. The most important change was an expansion of the number of performance indicators used by a system. Some initial systems (e.g., Minneapolis)107 used only citizen complaints. Others used citizen complaints and use-of-force reports. As time passed, there was increased recognition that one or two indicators are not adequate for identifying officers with potential performance problems. (This issue is discussed in greater detail in Chapter Four). Citizen complaints, for example, are filed in only a small number of potential cases, making them a very arbitrary indicator of officer performance.108 At the same time, it is increasingly recognized that other indicators–such as resisting arrest charges, use of sick leave–are very useful for identifying officers with potential performance problems. An Emerging Best Practice By the year 2001, EI systems had emerged as a recognized best practice in police accountability. A series of actions by the U.S. Department of Justice contributed to this development. A 1996 DOJ conference on Police Integrity recommended that police departments explore EI systems as an accountability mechanism.109 A 2000 conference sponsored by the COPS Office included a discussion of EI systems.110 Finally, the Justice Department produced the 2001 report entitled Principles for Promoting Police Integrity. Other recommended best practices include a comprehensive use-of-force reporting system, an open and accessible citizen complaint process, and data collection on traffic stops.111 In 1997, the Civil Rights Division of the DOJ entered into the first consent decree under Section 14141 of the 1994 Violent Crime Control Act. The consent decree with Pittsburgh included a requirement that the police department implement an EI system, which became known as the Performance Assessment and Review System (PARS). Subsequent consent decrees and memoranda of understanding have contained similar requirements. The consent decree with Los Angeles, for example, requires the LAPD to include seventeen performance indicators in its revised TEAMS II system.112 The memorandum of understanding with Cincinnati requires the police department to include ten performance indicators in its Risk Management System.113 Finally, in 2001, the Commission on Accreditation for Law Enforcement Agencies (CALEA) adopted Standard 35.1.15 requiring a Personnel Early Warning System.114 This recommendation by CALEA is particularly significant since the Commission represents the leaders of the law enforcement profession. In short, by 2001 a consensus of opinion had emerged among the U.S. Justice Department, federal courts, the law enforcement accrediting agency, and civil rights agencies that EI systems were a necessary part of a comprehensive approach to police accountability. Implementing EI Systems: A Major Challenge Although the concept of EI systems has become increasingly popular, the history of particular systems indicates that they are often extremely difficult to implement and maintain. When law enforcement agencies began to develop the first EI systems they were forced to do so without guidance from any published literature on the subject. There were no descriptions of systems, recommended models, or evaluations. Some departments simply devised their own systems, while others borrowed from other early systems.115 With the benefit of hindsight, it is now clear that implementation is a major issue with respect to EI systems. (This issue is discussed in detail in Chapters Three and Four of this report.) The NIJ evaluation of EI systems in three cities concluded that EI systems extremely complex administrative tools that require considerable administrative attention.116 An effective EI system requires careful planning and close, ongoing administrative attention once it is established. The problems associated with the development and implementation of an EI system are illustrated by the histories of several large departments. The Los Angeles Sheriff's Department PPI System The Personnel Performance Index (PPI) in the Los Angeles Sheriff's Department (LASD) is widely regarded as the most sophisticated EI system in the country. One report described it as "without question, the most carefully constructed and powerful management tool for control of police misconduct currently available in the United States."117 Representatives from the LASD, for example, have been invited to describe the PPI at national level conferences on police accountability. The history of the PPI, however, indicates the difficulties in establishing a first-rate EI system. The idea for the PPI originated with the Kolts Report on the LASD in July 1992. The report pointed out that the LASD did not have an effective mechanism for identifying officers who posed a high risk to the department with regard to civil liability claims. One of the main reasons was that the department did not systematically collect and retain relevant performance data. Many important items, for example, were maintained on paper 3x5 cards, and some of these records were periodically discarded. In order to achieve an effective risk management mechanism, the Kolts Report recommended the creation of what eventually became the PPI.118 The LASD responded favorably to the Kolts recommendation and planned to have the PPI operational by 1993. Because of a number of delays–similar to those that have affected other agencies-the target date for a fully operational PPI was pushed back to 1997. During this period, Los Angeles County settled a civil claim against the Sheriff's Department by promising to pay an additional $500,000 if the PPI was not fully operational by March 31, 1997. When that date arrived, the Special Counsel to the Sheriff's Department served as reviewer and concluded that the PPI system was indeed fully operational.119 In short, it took almost five years for the LASD to implement its EI system. This occurred in the context of strong support from both the Sheriff and the County Board of Supervisors, and in an agency that is technologically sophisticated. Nonetheless, in early 2003 the Special Counsel to the LASD delivered a very critical report on the PPI. A number of administrative problems had weakened the effectiveness of this potentially outstanding system. In particular, staff cutbacks had made it difficult for the LASD to enter performance data into the PPI database, with the result that commanders were not able to conduct timely analyses of officers' performance. The Special Counsel also found that many supervisors in the department were not familiar with the PPI and its capabilities. Some district commanders had spent time and money developing their own local systems because they were not aware that the data they wanted was readily available through the PPI. The Special Counsel's report clearly indicated that for an EI system to be fully effective, a department needs to maintain adequate staffing for the system and to engage in a continuous process of training officers–and commanders in particular–on the nature and capabilities of the system.120 The Los Angeles Police Department The story of the TEAMS in the Los Angeles Police Department (LAPD) represents a more disturbing picture of implementation problems associated with EI systems.121 It begins with the highly publicized beating of Rodney King on March 3, 1991. The Christopher Commission, appointed to investigate problems within the LAPD, found both that a small group of 44 officers had particularly serious performance records that were readily identifiable through existing departmental records, and that the department took no effective action to either discipline or correct the behavior of these officers. To address this problem, the Commission recommended the creation of an EI system.122 The LAPD's EI system became known as TEAMS (Training Evaluation and Management System). As a report by the Los Angeles Police Commission documents, development of the system over the next few years was erratic at best. In 1996, the Police Commission hired Merrick Bobb (then also the Special Counsel to the LASD) to conduct a five-year update on implementation of Christopher Commission recommendations. Bobb reported that the LASD did not have an operational EI system as intended by the Christopher Commission and that TEAMS was at that point "weak and inadequate."123 The president of the Los Angeles Police Commission then initiated a request for a federal grant to facilitate the implementation of an expanded version of TEAMS. The U.S. Department of Justice awarded a grant of about $175,000 in September, 1997, and the new EI system became known as TEAMS II. In 1999, the Rampart scandal embroiled the LAPD in further controversy, prompting three separate investigations of the department (one by the LAPD itself,124 one by the L.A. Police Commission,125 and one sponsored by the police union126). The Police Commission's report documented the history of the TEAMS II system and the fact that it was not currently operational. In fact, the city had not even drawn on the federal grant awarded to implement TEAMS II. Meanwhile, the U.S. Justice Department sued the City of Los Angeles and the LAPD under the 1994 Violent Crime Control Act, alleging a pattern or practice of abuse of citizens. The suit was settled through a consent decree that closely resembled similar decrees and memoranda of understanding negotiated by the Justice Department. The consent decree mandated the LAPD to complete the planned implementation of TEAMS II and specified that it should incorporate 17 separate performance indicators.127 In short, eleven years after the 1991 Christopher Commission recommendation, the LAPD did not have a fully operational EI system. Pittsburgh Police Department The Performance Assessment and Review System (PARS) in the Pittsburgh Police Bureau represents a case of a successful implementation. The system originated with a 1997 consent decree settling a U.S. Department of Justice suit against the department for failure to address excessive force problems. The Pittsburgh suit was the first to be settled under the provisions of Section 14141 of the 1994 Violent Crime Control Act authorizing the DOJ to bring suit where there is a "pattern or practice" of violating citizens' rights.128 The Pittsburgh consent decree has served as a model for subsequent agreements in Section 14141 litigation. Later agreements are very similar in overall content, but tend to be much longer and more detailed than the Pittsburgh agreement.129 The Pittsburgh PARS includes data on eighteen performance indicators. Officers are identified by the system if their number of indicators is one or more standard deviations greater than their peer officers–thus, night shift officers are compared with other night shift officers. One of the most significant aspects of PARS is that it is incorporated into the routine supervision process in the department. Supervisors are directed to log on to the system each day prior to roll call. As a result, each supervisor is able to note each and every activity that might require supervisory attention. The incidents are not limited to those officers who have been previously identified by PARS. A Vera Institute evaluation characterized the PARS process in terms of a "sweeping change in the duties of the lowest level supervisors."130 Implementation of the Pittsburgh PARS has been subject to two separate monitoring and evaluation procedures. The consent decree required the appointment of an independent monitor to oversee implementation of the decree and to file semi-annual reports with the court. The monitor found the Pittsburgh police department to be in compliance with the decree, including the provisions related to PARS, and in the summer of 2002 the court lifted the decree (for the police department but not for the independent Office of Municipal Investigations which is responsible for handling citizen complaints).131 Meanwhile, the consent decree and its implementation were evaluated by the Vera Institute of Justice. The evaluation, completed in 2002, found not only that the consent decree was successfully implemented, that it had a positive effect on many aspects of the police department, and that a number of alleged unintended consequences did not in fact occur. The Vera evaluation found that the EW system was functioning effectively as intended.132 The successful implementation of the EW system in Pittsburgh was due in large part to what the Vera Institute described as a "dogged determination" on the part of the police chief and other city officials to comply with the terms of the consent decree.133 The lesson of the Pittsburgh experience is that a sophisticated EI system can be successfully implemented in a large police department in a relatively short period of time, given sufficient leadership and resources. Research on the Effectiveness of EI Systems There are five sources of evidence on the effectiveness of EI systems. They include: (1) the NIJ study of three major police departments, (2) an evaluation of the consent decree mandating the creation of PARS in the Pittsburgh Police Bureau, (3) the reports of the Special Counsel to the Los Angeles Sheriff's Department which include periodic assessments of the LASD PPI, (4) anecdotal evidence from other departments, and (5) the UNO/PERF survey of police managers. The first three are discussed below. The UNO/PERF survey is reported in detail in Chapter Four. (1) The NIJ Study of Early Warning Systems The National Institute of Justice (NIJ) study was conducted by the author of this report and involved the EW systems in three large police departments: Miami-Dade, Florida; Minneapolis, Minnesota; and New Orleans, Louisiana. The study found all three EI (the earlier NIJ report used the term Early Warning, but this report uses the now-preferred term Early Intervention) systems to be effective in reducing citizen complaints and use of force among officers subject to intervention.134 The NIJ study included both qualitative and quantitative components and a number of previously unacknowledged issues were identified in the course of the evaluation. In many respects, the qualitative findings on the administration of EI systems are more significant that the quantitative findings on the impact of EI system intervention on individual officers. Most important, the study found that EI systems are extremely complex administrative tools that require close ongoing administration. Additionally, the authors concluded that it was difficult to separate the process and impact of an EI system from the larger climate of accountability within a police department. Qualitative Analysis of the Three EW Systems Each of the three police departments had very different histories with respect to accountability in the years immediately preceding the evaluation, and these histories affected the nature of their respective EI systems. The Miami-Dade police department had been in the process of enhancing accountability for almost twenty years. The department had a comprehensive personnel database in place that served the EIS but was independent of it. The Minneapolis Police Department, meanwhile, was in the process of raising its standards of accountability, particularly with respect to the use of force. The current chief was appointed in large part because of criticism of the lax disciplinary policies of the previous chief. The New Orleans police department, meanwhile, was also under the direction of a new chief executive who had been appointed to reform the department following a number of force and corruption scandals that generated considerable national publicity. The NIJ study concluded that the EI systems in each of the three departments were integral parts of ongoing departmentwide accountability efforts. In such a context, it is difficult to specify the independent effect of the EI system separate from other reforms that occur in the department generally. As has already been pointed out, the qualitative research also found that EI systems are far more complex administrative mechanisms than is generally recognized. The study found that EI systems require enormous administrative attention if they are to be implemented and maintained effectively. The study dispelled the notion that an EI system is simply a tool (e.g., an alarm clock) that can be purchased off the shelf, plugged into the department, and then left to run by itself. (Chapter Three of this report provides even stronger documentation on the importance of planning and administrative follow-through.) Evidence of the administrative complexity of an EI system is illustrated by several examples drawn from the three departments studied. One of the EI systems had actually ceased operation for a period of about one year because of disruptions related to administrative changes in the department. One of the EI systems had undergone considerable revision and improvement since it was first launched. The officials in charge called their efforts "a work in progress." One of the EI systems studied had serious problems related to inputting data in a timely fashion. Two of the departments studied had rudimentary computer systems for their EI systems, such that it was possible to retrieve the files for individual officers but not to generate data on larger patterns and relationships between incidents and officers. Quantitative Evaluation The quantitative component of the evaluation attempted to assess the impact of each EI system on officers who were subject to departmental intervention. Because of the nature of the individual EI systems and the data available, two of the EI systems were subject to one mode of evaluation and the third subject to a different mode. In Miami-Dade and Minneapolis, the evaluation methodology consisted of an analysis of the performance records of all officers hired in certain years (1990-1992 in Miami-Dade; 1991-1992 in Minneapolis). Each cohort of officers was partitioned into those officers who had been subject to the EI system and their colleagues who had not been subject to it (non-EI). The two groups were then compared in terms of background characteristics and performance records. The performance records of the EI group were analyzed in terms of pre-intervention and post- intervention. The nature of the New Orleans PPEP, which involved a four-day class with a group of officers, offered a unique opportunity for a different methodology. First, one PPEP class was directly observed for the purpose of collecting data on the content of the intervention and the observed responses of officers in the class. Second, because all officers subject to PPEP classes are invited to complete a course evaluation form, it was possible to analyze the responses of all officers who had ever been subject to the PPEP intervention. Findings (1) Officer Backgrounds The data in Miami-Dade and Minneapolis found that race or ethnicity was not a factor in with regard to those officers subject to EI intervention. White, African American, and Hispanic officers were selected for EI in proportions closely related to their presence in the larger cohort. This finding is particularly significant because it does not support the popular view that the problem of police use of excessive force is primarily a matter of white police officers mistreating racial and ethnic minorities. It is consistent with other research indicating that race or ethnicity is not a determining factor with regard to law enforcement officer behavior.135 Female officers, meanwhile, were only about half as likely to be subject to EI intervention, as their presence in the larger cohort would have predicted. This finding is consistent with other research indicating that female officers are less likely to be involved in use of force incidents and are rarely identified as officers with patterns of repeated use of excessive force.136 (2) Performance Histories Officers selected for intervention by the EI systems in Miami- Dade and Minneapolis did in fact have significantly worse performance histories prior to intervention when compared to their colleagues. In Miami-Dade, for example, officers selected by the department's Early Intervention System (EIS) averaged twice as many use-of-force reports per year as non-EIS officers. EIS officers were twice as likely to have received four or more citizen complaints than their peers, were twice as likely to have received a formal reprimand, and almost three times as likely to have been suspended. In Minneapolis, officers selected by the department's EI system averaged twice as many complaints as their non-EI peers, were slightly more likely to have been reprimanded, and were three times as likely to have been suspended. In one troubling finding, EI officers in both Miami-Dade and Minneapolis were slightly more likely to be promoted than their non-EI peers, although in Miami-Dade the difference was not statistically significant. This issue merits further research. The data from this study does not permit investigation into the reasons for this finding. It is possible that for some officers, the negative aspects of assertive behavior that results in EI system identification also includes positive aspects that help earn promotion. Future research and policy development can, hopefully, permit EI systems to identify and curb the negative aspects of assertive behavior while retaining the positive aspects of officer assertiveness. This finding lends support to the anecdotal view, supported by investigative journalists' reports that a small group of officers in any department do have significantly worse performance records than their colleagues. (3) Impact of Intervention The evaluation found that in Miami-Dade and Minneapolis the EI systems had significant and positive impacts on officers' performance histories. Among officers subject to EI system intervention, citizen complaints and involvement in use-of-force incidents declined by about two-thirds. In Miami, 27 of the 28 EIS officers had at least one use-of-force report; following intervention only 14 had a use-of-force report. Twenty of the 28 officers had four or more complaints prior to intervention, while only nine of the 28 had four or more complaints following intervention. In Minneapolis, EI officers received an average of 1.9 complaints per year prior to intervention and 0.65 per year after intervention. One curious aspect of the findings is that the improvements in officer performances were virtually identical in Miami-Dade and Minneapolis despite the fact that the two EI systems had very different administrative histories. The Miami-Dade system had been in place for over 15 years and was well-integrated into the department. The Minneapolis system, on the other hand, was far less sophisticated (with regard to performance indicators) and had experienced some significant disruptions. One possible explanation is that an EI system works effectively with many (and perhaps most) officers simply by putting them on notice (e.g., warning) and that an elaborate intervention is not necessary. (4) Officers' Responses to Intervention The nature of the New Orleans PPEP made it possible to both observe officers in the PPEP class and to analyze responses to the course evaluation. This component of the evaluation yielded extremely important findings. First, the direct observation of the class found that officers responded in very different ways to different units and teaching methodologies. During the class, most of the officers expressed open hostility to the PPEP, referring to it as "politeness school," or "bad boys class." Officers were visibly disengaged from aspects of the course that involved traditional lectures and in particular lectures about general topics such as the role of the police, stress management, and the problem of substance abuse in American society. Some officers appeared to be close to falling asleep in the class during these units. Some expressed their disinterest through obvious body language. In sharp contrast, however, they were actively engaged in the unit of the course that asked them to critique specific police-citizen encounters. The case studies were drawn from the department's files, and in a few cases the officers claimed to remember the incident (none involved officers in the class). In this exercise, the officers appeared to be knowledgeable and professional police officers, fully aware of proper procedure in difficult situations, and proud of their own knowledge about how to control situations effectively. It should be noted that this engagement was exhibited by the same officers who had earlier voiced extremely hostility to the PPEP. The lesson of this finding is that interventions that involve traditional lectures are not likely to be effective in engaging officers about performance issues. The use of active learning techniques that allow officers to draw on their experience and knowledge, on the other hand, appear likely to engage their attention. In the observed case, however, it is not known whether the officers in the class internalized the lessons from the case studies and altered their own behavior accordingly. With respect to the evaluations of the PPEP classes, officers who completed the evaluations gave the class an extremely positive rating. On a scale of 1-10, they gave it an average rating of 7.0. When the four officers who gave it a rating of one (the lowest possible rating) are omitted, the average rating rises to 8.1. These responses are extremely significant in light of the fact that the officers in the observed class openly disparaged the PPEP program. The positive responses on the numerical evaluation were supported by the narrative comments on the evaluation forms. All of the officers made at least one positive comment about the class. Comments included favorable references to the verbal judo/complaint reduction component and the stress reduction component. About three-quarters of the officers also made negative comments. Significantly, however, all the negative comments were about the PPEP in general or the department and its leadership. There were no negative comments about the PPEP class per se. The positive evaluations of the PPEP classes are an extremely important finding. It suggests that most officers appreciate practical assistance designed to help improve their performance. This finding is reinforced by the PERF survey of police managers (See Chapter Three) which found no major hostility to EI systems from rank and file officers. Limitations of the Study While extremely encouraging, the evaluation was subject to a number of limitations that need to be noted: First, this was a retrospective study utilizing available official data. Consequently, it was subject to the limitations of that official data. While the study sought to control for important variables, the investigators are not 100 percent certain that all the data are accurate. One particularly important issue involves officers' assignment. Reassignment to a low-crime area could account for a significant drop in citizen complaints and use-of-force incidents. Official records with regard to assignment histories, while good, are not necessarily perfect. Prospective evaluations in which investigators can monitor assignment patterns and other relevant factors can reduce or even eliminate these problems. Second, the design of this study does not make it possible to determine exactly what aspect of the EI intervention accounts for the positive results. It is entirely possible that for many officers simply being selected by the EI system was a sufficient warning that led to improved performance. A number of police officials involved with EI systems believe that this is true. The effect here is independent of the actual content of the intervention. It is equally possible, however, that some officers benefited from specific aspects of the intervention: advice on how to de-escalate confrontations, or to handle groups of troublesome citizens. It is entirely likely that different officers respond differently. Future research on EI systems should focus on specifying what kinds of intervention are appropriate and effective for different kinds of officers. It may also be useful to think in terms of triage: that some officers may need only a warning, that others may need help with specific performance issues, and that some may not respond at all to intervention. (2) Evaluation of the Pittsburgh PARS The Pittsburgh PARS has been evaluated in two different ways. First, the consent decree included a court-appointed monitor to oversee implementation of the decree. The reports of the monitor found that the Pittsburgh Police Bureau successfully complied with the terms of the decree, including implementation of the PARS. The decree was lifted in the summer of 2002. Second, the Vera Institute and PARC conducted a general evaluation of the impact of the 1997 consent decree settling a U.S. Justice Department pattern or practice suit against the department. The evaluation found that PARS had a generally positive impact on the department. Several supervisors testified that it did enhance the quality of supervision in the department. At the same time, they expressed concern about the burden of paper work required by PARS. One supervisor offered an unverified estimate that the amount of time sergeants devoted to traditional street-level supervision declined from 70-80 percent of total time to 25-30 percent.137 An important aspect of the Vera/PARC evaluation was an analysis of some departmentwide activity data (although trends in these data are not necessarily attributable to PARS alone). The data found, for example, that despite anecdotal reports of a serious decline in officer morale because of the consent decree, the number of sick days taken by officers actually declined from 1995 to 1999. Also, there was no upsurge in separations from the department between 1996 and 1999, despite anecdotal reports to the contrary. The data indicated a decline in arrests beginning in 1994, but this trend paralleled a decline in reported crime and could not be attributed to the consent decree. Finally, there was no decline in traffic tickets issued following the 1997 consent decree (a sharp decline occurred in 1996 as a result in a change in the rules for overtime pay). (3) Assessment of the LASD PPI System The Special Counsel to the Los Angeles Sheriff's Department reports data indicating that the PPI is very effective in achieving its goals. The Special Counsel a form of external citizen oversight for the LASD and operates through a contract with the Los Angeles County Board of Supervisors. The Special Counsel reported monthly average data on officers subject to Performance Review (PR) for periods three years prior to PR, the period during PR, and the period after PR. The rate of officer-involved shootings dropped from an average of .50 per month before PR to zero following PR. Use-of-force incidents dropped from an average of 7.11 per month before PR to .98 per month after PR. And personnel complaints dropped from an average of 3.86 to .74 in the same time periods.138 One interesting finding was that 11 percent of all officers placed on Performance Review left the department within a short period of time. This included ten retirements, six discharges, one resignation, and one departure for reasons that were not clear. In short, two-thirds were voluntary departures.139 To the extent that the PPI encourages problem officers to leave the department voluntarily, it saves the department both the direct expense of conducting a full two-year Performance Review and the risk that an officer will engage in other serious misconduct. The 16th Semiannual Report of the Special Counsel, however, was very critical of the PPI. As noted earlier, it found that the PPI had not been fully staffed and as a result performance data were not being entered into the database in a timely fashion. Also, many commanders were not aware of the nature and capabilities of the system.140 As a result of the Special Counsel's report, LASD commanders made a public commitment to make the necessary changes to ensure that the PPI was fully effective and capable of achieving its basic goal of reducing litigation costs to the county. The Special Counsel's report highlighted the administrative demands of an EI system and the need to continuously oversee the operation of a system. (4) Anecdotal Evidence From Other Departments At the Early Intervention State of the Art Conference in early 2003, commanders from several EI systems were able to describe case studies of individual officers who had been helped by EI intervention. One commander, for example, described the case of a female officer who was not engaging citizens in a properly assertive manner, a practice that encouraged aggressive behavior on the part of some citizens (and resulting uses of force by the officer). Intervention counseling determined that the officer had a great fear of being struck in the face. Training helped the officer become skilled in the proper techniques for engaging potentially aggressive citizens. As a result, the officer's performance improved dramatically. The commander from another department described the case of an officer whose performance was being affected by serious personal financial difficulties. Intervention counseling helped the officer obtain professional financial advice and his performance subsequently improved.141 The two cases cited above are anecdotal and not sufficient to validate the overall effectiveness of an EI system. Discussions at the EI State of the Art Conference explored the possibilities of developing more systematic ways of documenting individual success stories (with personal identifiers eliminated to preserve privacy). (5) The Police Managers' Survey The findings of the police managers' survey, which includes both managers experienced with EI systems and managers without such experience, is reported in Chapter Four of this report. Conclusion EI systems originated as administrative tools for identifying a few officers with performance problems. They have evolved into more elaborate tools for reviewing officer performance and achieving higher standards of accountability. Research to date indicates that EI systems can be effective in achieving their goals. At the same time, the history of EI systems in several departments indicates that EI systems are difficult to implement effectively. Careful planning and continuous administrative oversight is necessary. Chapter Four Lessons From the Field: Police Managers Perceptions and Experiences With Early Intervention Systems The PERF Survey....................73 Part One: Experiences With Early Intervention Systems .............................74 Source of their EI Systems ........................................................76 Overall Assessment of the System...................................................76 Impact on the Quality of On-the-Street Police Service..............................76 Positive Contributions to Supervision and Accountability...........................76 Strengthens the Role of Supervisors ...............................................77 Early, Non-Disciplinary Intervention ..............................................78 Ability to Track Officers Who Change Assignments...................................78 Problem-Solving Supervision........................................................78 Timely, Objective Data ............................................................79 Lack of Opposition From Rank and File and Unions ..................................79 More Positive Than Negative Impact on Officer Morale ..............................80 Changes in Officers’ Attitudes ....................................................80 Differential Impact on Officers ...................................................81 No Problems Regarding Confidentiality .............................................81 Problems With EI System............................................................81 Intervention SystemsImplementation Problems........................................82 Lack of Commitment by the Deparment ...............................................83 Infrastructure Needs...............................................................83 Successful Planning ...............................................................83 Need for Continued Revision and Improvement........................................83 What Needs to be Changes or Improved...............................................84 Part Two: Departments Without Early Intervention Systems ..........................84 Managers Knowledgeable About EI Systems............................................84 Managers Overwhelmingly Anticipate Positive Impact of an EI System.................85 Managers Anticipate Objective Systematic Data......................................85 Managers Anticipate Positive Impact on Supervisors.................................86 Managers Anticipate a Reduction in Lawsuits Against the Department.................86 Managers Anticpate Multiple Benefits...............................................86 Anticipate a Positive Impact on Community Relations ...............................87 Overestimate the Opposition From Within the Department.............................87 Understand Opposition Will Come From the Officers With Performance Problems .......87 Overestimate Union Opposition......................................................88 Underestimate Administrative Problems..............................................88 Selling the Idea: Benefit to Subject Officers......................................88 Selling the idea: Makes the Supervisor’s Job Easier ...............................89 Recognize Limitations of Current Accountability Mechanisms ........................89 Able to Track Officers Who Change Assignments .....................................90 Reactive Versus Proactive..........................................................90 Prefer Information From Other Law Enforcement Agencies ............................90 Conclusion ........................................................................91 Lessons From the Field: Police Managers’ Perceptions and Experiences With Early Intervention Systems After more than a decade of experimentation and development (and in some departments almost two decades), law enforcement agencies have a considerable body of practical experience with Early Intervention systems. To tap into this experience, we surveyed law enforcement managers. As described in detail below, this survey used innovative online techniques, surveying members of the Police Executive Research Forum (PERF). The methodology and principal findings are reported in the following sections. The survey used the term Early Warning, and consequently that term is often used in this chapter. The PERF Survey This survey represents collaboration between the University of Nebraska at Omaha, the Police Executive Research Forum (PERF), and DiscoverWhy. Funding was provided by the Office of Community Oriented Policing Services (COPS) of the U.S. Department of Justice. DiscoverWhy is a private survey research firm developing innovative online technologies for surveying public opinion. This project represents an experiment in applying these technologies to law enforcement and criminal justice research. PERF is a professional association of police managers, with a membership of nearly 600 commanders and supervisors across the country. PERF has been involved in cutting edge research, training and demonstration projects since 1975. The COPS Office provided a grant to the Criminal Justice Department of the University of Nebraska at Omaha to provide technical assistance to law enforcement agencies regarding Early Warning (EW) systems. This survey represents one component of that project. PERF provided a list of email addresses for its members. Each member then received an email invitation to participate in the survey. Those individuals indicating they wished to participate then received a password allowing them access to the survey instrument. Participants could then log on and participate at their convenience. The DiscoverWhy technology permits an electronic survey, with both closed and open-ended questions, that may be accessed and completed at the convenience of each participant. The open-ended questions are particularly valuable for tapping qualitative assessments of the subject (in this case, EI systems) and in the process eliciting valuable data embedded in mixed and ambiguous responses. PERF submitted a mailing list of 831 names. Email invitations were sent to 521 members with currently active email addresses. A total of 135 members participated in the survey, for a response rate of 26 percent. About 40 percent of the respondents work in departments with EI systems, while 60 percent work in departments without EI systems. This distribution is consistent with the findings of an earlier survey of EI systems.142 Part One: Experiences With Early Intervention Systems Law enforcement managers who have had practical experience with an EI system report very positive assessments of its impact in their departments. They report that it has had a positive impact on supervision and accountability and on the quality of on-the-street police work. Especially important, they have not experienced serious opposition from rank and file officers or their police unions. Many, in fact, report that the attitudes of rank and file officers changed from suspicion to understanding and support for the EI system. Experienced managers also report a number of problems with EI systems, however. The most important problems fall into two general categories: a failure to communicate the nature and purpose of these systems during the planning stage and a failure of some command officers to follow through on their responsibilities once the system begins operating. In short, these are management problems and not problems inherent in EI systems. General Assessment of EI System Image of a bar showing showing the general assessment of EI System Impact of EI System on Quality of Police Service Image of a bar graph showing the impact of EI System on Quality of Police Service Source of their EI System Most departments (70 percent) developed their own EI systems, while 20 percent borrowed theirs from other departments. Overall Assessment of the System Experienced police managers report overwhelmingly positive experiences with their EI systems: 64 percent reported a positive general assessment and 32 percent reported a mixed general assessment. The nature of these mixed assessments is explored in responses reported below. Impact on the Quality of On-the-Street Police Service Managers overwhelmingly reported that the EI system had at least some positive impact on the quality of on-the-street police service. Almost half (49 percent) reported a positive impact, while 28 percent reported a mixed impact. This represents a combined total of 77 percent of respondents who report at least some positive impact. Perhaps most important, no respondents reported a negative impact on the quality of on-the-street police service. In particular, no manager reported that the EI system caused officers to back off and reduce their activity level. Positive Contribution to Supervision and Accountability Managers report that their EI systems have made positive contributions to supervision and accountability in their departments. Comments from these managers include: • The process assisted us in identifying officers or non-sworn personnel who have field responsibilities who began to demonstrate behavior not consistent with our policies and standards. • Raises the awareness of lack of proper supervision. • Supervisors pay more attention to what is going on, document all activities more thoroughly, and talk with officers when things might look like a problem. Early, Non-Disciplinary Intervention Mangers report that, consistent with the objectives of EI systems, the EI system has been an effective instrument for early, non- disciplinary intervention with officers who exhibit performance problems. Two managers, for example, reported that: • "We have been able to intervene with help before misconduct occurs that requires discipline." • "[The] program provides a way for the department to provide non-disciplinary direction and training before the officer becomes a liability to citizens, the department, and him/herself" Ability to Track Officers Who Change Assignments One problem in police departments is that because of regular changes in assignment many supervisors do not know the performance histories of new officers under their command. An EI system database overcomes this problem. • There is a lot of movement of personnel, so supervisors often do not know the histories of their officers. The EWS report brings them up to speed in a much more timely fashion. • No one falls through the cracks any longer. Problem-Solving Supervision Some managers understand that an EI system facilitates a problem-solving approach to supervision. • A useful tool to involve supervisors and lieutenants in non-traditional model of problem-solving. It has served to enhance their management skills and help round out their people interaction skills. Timely, Objective Data Managers report that having objective performance data helps to eliminate charges of favoritism, levels the playing field since no one can be accused of playing favorites, and makes it much easier to deal with the police union. Lack of Opposition From Rank and File and Unions The most surprising finding is that managers report that they have not experienced opposition to the EI system from their local police union. This finding is particularly significant because managers who do not presently have an EI system in place anticipate such problems (see below). Eighty-four percent of the managers with EI systems report having had no problems with their unions regarding their systems. Problems With Police Union? Image of a bar graph showing the percentage of problems with police unions • As command staff members, we are constantly looking for ways to pick up on officers in need of assistance. The Early Warning System has picked up on several officers in the department who were headed down the wrong path Strengthens the Role of Supervisors Police managers report that their EI systems strengthen the role of supervisors. Some of their comments include: • A useful tool to involve supervisors and lieutenants in a non-traditional model of problem solving. It has served to enhance their management skills and help round out their people interaction skills. • Sergeants have been able to evaluate the strengths and weaknesses of their squads even before meeting with them, and therefore are able to develop proactive strategies to address personnel and leadership issues. • When an officer is placed in the program, for whatever reason, his or her supervisor is also involved in the process. The Early Warning System Board also looks at the actions of the supervisor to determine if the supervisor is providing the supervision needed to correct the problem. Impact on Management and Supervision Image of a bar graph showing the impact of management and supervision One of the few managers who encountered problems with the union stated: "The union believes that the EWS is a witch hunting tactic, used by the administration to target disliked officers. In fact, no one has been disciplined because of EWS, it just wasn't explained very well." More Positive Than Negative Impact on Officer Morale Twice as many managers (12 percent) reported that the EI system had a positive impact on morale among rank-and-file officers as reported a negative impact (six percent). Slightly more than half (52 percent) reported that the EI system had a mixed impact on officer morale. Finally, 30 percent reported that the EI system had no impact on rank and file morale. Impact on Officer Morale Image of a bar graph showing the impact on officer morale Explaining the positive impact on morale, one manager pointed out that "nothing hurts morale like no action being taken against problem officers. It shows supervision that problem people can be dealt with." Changes in Officers' Attitudes Managers report changes in rank and file officers attitudes, from suspicion or hostility to growing support for EI systems. • The system was met with much cynicism and distrust when first introduced. Experience and education has reduced those attitudes. • Most no longer believe this system is out to "get" them, but rather to assist them. Differential Impact on Officers Some mangers reported that the impact of the EI system varied from officer to officer. They reported, for example, that: • Some officers take it to heart, others see it as unfair discipline. • The solid officers of the department are supportive of the system both conceptually and practically. No Problems Regarding Confidentiality Virtually no problems were reported with regard to breaches of confidentiality, such as improperly leaking information about which officers had been identified by the EI system. In fact, the only reported case involved a situation where the officer subject to intervention disclosed that information himself. Problems With EI System The overwhelming majority (85 percent) reported that they had experienced some problems with their EI system. Only 15 percent said they had had no problems. Problems With EI System? Image of bar graph showing problems with EI System Implementation Problems Mangers did report a number of problems with their EI systems. Virtually all of these problems were related to implementation issues, in particular a lack of communication about the nature of the EI system and a lack of follow-through on the part of responsible officials. The greatest number of problems involved a failure to communicate the nature and purpose of the EI system to officers in the department. Commanders identified this problem in response to several different questions on the survey. Some commanders explained, for example: • [Top management] did not explain the purpose of the program well [to mid-level supervisors]. • [Departments needed to deliver better] training and introduction to the first line supervisors. • [The agency] could have done a better job of pre-selling and training on the benefits of an EWS. • While the system was crafted appropriately, it was not explained to the officers or the first line supervisors to the extent necessary to make it an understandable and viable system. • The system is misunderstood by the first line supervisors, and therefore retraining is needed at the application level. Lack of Commitment by the Department Another problem reported by experienced managers is that while the EI system is good in principle, the department as a whole and some mid-level supervisor did not follow through with their responsibilities, thereby undermining the effectiveness of the system: • The program does not hold a prominent place in the organization. As a result, no results are provided to commanders as to its effectiveness or benefits. • Our program lacks real involvement from the commanders. Typically the officer's actions are justified or rationalized by the commanders and rarely is there a consequence for the officer. • [There was] irregular review and lack of consistent follow up. Infrastructure Needs The lack of an adequate technological infrastructure was also a major problem. About 43 percent of the managers reported that their department did not have the necessary infrastructure in place when their department began planning its EW system. Successful Planning Many managers reported that their EI planning effort was successful. The key to success was the involvement of all relevant stakeholders: • Gathering input from all involved, officers through management as well as the union. • We involved Risk Management, the Deputies Union, and management. • Involving a wide range of personnel and viewpoints from the start. Need for Continued Revision and Improvement Several managers acknowledged that their EI systems were not perfect and that further revision and improvement was necessary. "The system has worked well. EWS needs to continuously evolve and expand." What Needs to be Changed or Improved? When asked what they thought needed to be changed in their EI systems, managers most commonly reported that communication problems needed to be addressed: 27 percent indicated that improved communication between command staff and rank-and-file was their most pressing need, while 14 percent said that it would be most important to improve communication among command staff. The need to clarify certain aspects of the EI system was also mentioned: 22 percent stated that it would be most important to clarify the relationship of the EI system to the department's formal disciplinary system, while 19 percent report that it would be most important to clarify the purpose of the EI system. These problems are also essentially communication problems. Part Two: Departments Without Early Intervention Systems About 60 percent of the managers surveyed work in departments that do not currently have operational EI systems. Their responses indicate their knowledge about EI systems and their perceptions of both potential contributions and problems. Managers Knowledgeable About EI Systems Generally, police managers in departments without EI systems are very knowledgeable about them. Only 13 percent of respondents said that they had no knowledge of EI systems, while another ten percent, based on specific comments they made, were misinformed about them. Thus, 77 percent of all managers had at least some knowledge about EI systems. Managers Overwhelmingly Anticipate Positive Impact of an EI System Managers overwhelmingly (94 percent) have a positive view of EI systems and anticipate that one would have a positive impact in their departments. Only three percent believe it would have a negative impact in their departments, another three percent have no opinion about the effect. Anticipated Impact of EI System Image of a bar graph showing the anticipated impact of the EI System Managers Anticipate Objective Systematic Data One of the advantages managers anticipate includes objective and systematic data: • EI systems allow the agency to use an objective scale by which to assess the need for intervention and action to prevent future problems. • EI systems store data in a single location, thus offering the potential to identify issues and problems in a timely fashion. • EI systems offer a systematic approach to identifying officers with performance problems. • EI systems allow supervisors to rely less on reputation and be more objective in performance evaluations. • EI systems instill more confidence that performance is being monitored and dealt with in a consistent fashion. • EI systems are consistent with what we [as the agency] already do, but would provide a more formalized system within which to do it. Managers Anticipate Positive Impact on Supervisors Managers anticipate that an EI system will help supervisors do their jobs. And as the previous section indicated, these expectations are generally confirmed in practice. • EI systems provide high-potential tools to help command staff members identify trends that they may miss. • EI systems help supervisors, middle managers, and administrators identify internal problems dealing with officer attitudes, performance, and skills. Managers Anticipate a Reduction in Lawsuits Against the Department Managers anticipate that an EI system will help to reduce lawsuits against the department. Significantly, however, this issue was not mentioned by managers with EI systems in place. They did, however, mention a reduction in complaints, including preventing future liability, enhancing officer retention, fewer personnel issues, and fewer law suits. Managers Anticipate Multiple Benefits A number of managers reported that they anticipate multiple benefits to their department from having an EI system. • EI systems help managers deliver employee assistance, risk management, improved morale, and improved employee and agency performance. • EI systems provide liability prevention, good community relationships, and good labor/management relationships. Anticipate a Positive Impact on Community Relations Managers overwhelmingly (82 percent) anticipate improvements in community relations. This factor was rarely mentioned by managers in departments with EI systems, however. Overestimate the Opposition From Within the Department Thirty percent of managers anticipate negative responses in their department if they implemented an EI System and 40 percent anticipate a mixed response. Thirty-seven percent anticipate opposition from rank and file officers, for example, and 29 percent would anticipate opposition from the police union. Another 30 percent anticipate a positive response. This represents a far more pessimistic view than has proven to be the case in departments with EI systems as reported in the previous section. Managers anticipate: • Some initial reluctance by supervisors and some uncertainty by officers on the goals of the EI system. • Skepticism at first, but acceptance that the system will be deployed. • Mild grumbling by the rank and file, minor hand-wringing by mid-management, brief tooth gnashing by the chief, followed by the system being implemented, rarely triggered, and quickly assimilated as just another obscure general order in the manual. • Skepticism until the benefits are proven. Understand Opposition Will Come From the Officers With Performance Problems Many mangers that anticipate opposition from the rank and file, however, recognize that most of it will come from the few problem officers. Managers anticipate: • Negative feedback from the bargaining unit, positive feedback from the commanders, but negative feedback from the small percentage of police officers that cause the most problems. Ninety percent of the officers would understand and accept the system. • Employees would accept it unless they thought they might be personally impacted. One respondent said, "Those are the ones that we need it for." Overestimate Union Opposition Many managers overestimate the extent of union opposition. Managers with experience, however, have not found significant opposition (See Section One). Often, opposition manifests in unexpected places. In practice, administrative problems tend to overshadow most other issues. Underestimate Administrative Problems In light of the problems reported by managers with EI system experience, managers without experience underestimate potential administrative problems. Only 29 percent mentioned administrative problems as the major obstacle they expected to face if implementing an EI system. Prospective EI implementation planners anticipate: • That the main obstacle will be collecting and displaying the data. • Inadequate staffing to run the system. • Concerns about the duplication of work, specifically how it will co-exist with the current supervisory and complaint tracking system. Selling the Idea: Benefit to Subject Officers When asked how they would sell the idea of an EI system to their department, almost half (47 percent) mentioned that it would actually benefit officers who are subject to the system. Prospective EI system planners anticipated: • "I would be able to point to specific examples that would demonstrate that if the agency had such a system in place, careers may have been saved." • "It permits early intervention to assist officers with problems before job security becomes an issue." • "We are trying to save careers and quite possibly families. I don't think it will take a 'sales job.'" Selling the Idea: Makes the Supervisor's Job Easier Many managers anticipate that they can sell the idea by emphasizing its proactive problem-solving capacity. Some managers' responses to the survey included: • "It would make their job easier. A good system could help in the early identification of problematic officers and procedures." • "Allows us to spot problems before they become career- threatening, eases workload by allowing the EWS to note where effort should be applied rather than waiting on a supervisor to recognize the trend." • "To keep your subordinates out of harm's way makes the supervisor's job easier." Recognize Limitations of Current Accountability Mechanisms In describing the potential benefits of an EI system, many managers acknowledged the limitations of the current accountability mechanisms in their department. Only 33 percent rated their departments' current accountability measures as "very effective." About half (47 percent) rated their departments' accountability measures as "adequately effective." These limitations include a lack of consistency, the absence of data or objective standards, and excessive discretion in handling alleged misconduct. Some of the limitations managers reported included: • "[The EI system is] not consistent, no standard system in place. Only a few supervisors take it seriously." • "The Professional Standards Unit is responsible for communicating Internal Affairs' (IA) role to Administration. This works for identifying recurring problems. The limitation is that it relies upon one supervisor and some complaints may not get as high as Internal Affairs." • "Minor complaints are most often handled by first line supervisors and are not documented in any retrievable form over time." • "Lack of guidelines leaves a lot of discretion with individual supervisors and therefore inconsistency across the department." • "It is informal and relies on the sergeant's ability to perceive problems and communicate effectively." Able to Track Officers Who Change Assignments Managers recognize that an EI system can help to track the performance of officers who change assignments and to provide new supervisors with useful data on their performance histories. • Officers sometimes switch assignments frequently and get a fresh start with new supervisors. • EI systems help track officers across different shifts. • Traditionally, officers' reputations tend to be directed by rumor and perception only. Reactive Versus Proactive Traditionally, supervisors respond to officers' performance reactively rather than proactively. • Supervisory action traditionally occurs after the fact and after discipline has been imposed several times. • Supervisory action traditionally is ad hoc and sometimes the conduct rises to the level of formal action before we initiate corrective action. Prefer Information From Other Law Enforcement Agencies Managers expressed a preference for information about EI systems from other law enforcement agencies. • 47 percent of respondents stated that if they needed additional information about EI Systems, the best source of that information would be expert personnel from another department with an EI system. • 30 percent of respondents stated that the best source of that information would be printed material. • 20 percent stated that if they needed additional information about EI systems, the best source of that information would be an International Association of Police Chiefs (IACP) training conference Conclusion Police managers who have experience with EI systems report very positive experiences with them. These managers have found EI systems to be an extremely useful tool for achieving accountability, with many benefits for supervisors and rank and file officers. This testimony is particularly important, as this survey represents the first systematic investigation of practical experiences with EI systems. They do not report strong opposition from rank and file officers or police unions, despite the fact that these problems are anticipated by managers in departments without EI systems. Managers do, however, report a number of problems related to the planning and implementation of their EI systems, particular with regard to communicating the nature and purpose of the system to all members of the department. Significantly, managers in departments without EI systems underestimate the management problems they will face. Chapter Four presents a guide for planning and implementing an EI system. The guide is based in part on the findings reported in this chapter. Chapter Five A Planning Guide for Implementing an Early Intervention System About This Chapter ............................................................95 Planning Process and Key Issues ...............................................96 STEP ONE: Recognize the Need ..................................................96 STEP TWO: Involve the Command Staff Early .....................................96 STEP THREE: Create a Planning Process .........................................97 STEP Four: Pinpoint Responsibility ............................................98 STEP FIVE: Involve Officers From All Ranks.....................................98 STEP SIX: Utilize Outside Experts .............................................99 STEP SEVEN: Assess the Department’s Computer Infrastructure....................99 STEP EIGHT: Keep the Staff Informed ..........................................100 STEP NINE: Keep External Stakeholders Informed................................100 STEP TEN: Identify and Address Specific Issues ...............................101 STEP ELEVEN: Take the Time to Get It Right ...................................105 STEP TWELVE: Provide Staff Training ..........................................105 STEP THIRTEEN: Choose Software ...............................................106 STEP FOURTEEN: Finalize the Plan .............................................106 STEP FIFTEEN: Implement the EI System.........................................106 A Planning Guide for Implementing an Early Intervention System About This Chapter Police managers experienced with EI systems strongly believe that the development of an EI system should involve careful planning. Many of the early EI systems made mistakes that had to be corrected later. Often, these were mistakes that could have been avoided. Also, any agency developing an EI system today has the benefit of these early experiences and can plan accordingly. This chapter is designed as a guide to help law enforcement agencies plan and implement Early Intervention (EI) systems. Specifically, it: ** Outlines the steps in a planning process ** Identifies the key issues in planning This chapter closely follows the planning process described in Samuel Walker, Carol Archbold and Leigh Herbst, Mediating Citizen Complaints Against Police Officers: A Guide for Police and Community Leaders (2002), published by the U.S. Department of Justice, Office of Community Oriented Policing Services (COPS). That report is available on the web at www.cops.usdoj.gov. Planning Process and Key Issues STEP ONE: Recognize the Need The first step in developing an Early Intervention system is to recognize the need. EI systems are now a recognized best practice in law enforcement, so there should be little room for debate on this question. But law enforcement agencies tend to have specific problems. It might be large civil litigation claims in one department, but repeated complaints about traffic enforcement in another. Some agencies took the first steps toward an EI system with committees to review abuse of sick leave or excessive motor vehicle accidents. Agencies planning to develop an EI system should begin by identifying any specific problems they have that should be addressed by the system as well as existing programs or data systems that could be incorporated into an EI system. Key Issue: Begin with a self-assessment, identify existing problems, current accountability efforts, and the potential value of an EI system. STEP TWO: Involve the Command Staff Early The senior command staff should be involved in the planning of the EI system at the earliest moment. These officials will be responsible for developing, implementing, and administering the system, and therefore it is essential that they fully understand and become committed to it. It is essential that the chief executive make it clear that the EI system is a top priority for the agency. The initial meetings with the senior command staff represent an opportunity for presentations by outside experts who can answer questions about such issues as the goals, operating procedures, potential advantages, potential problems, and administrative requirements of an EI system. Key Issue:The chief executive needs to make it clear that the program is a top priority and that staff members will be held accountable for failing to follow through on developing the program. STEP THREE: Create a Planning Process A formal planning process should be created. This involves the appointment of a planning committee with a clear mandate regarding the general outlines of the desired EI system. A timetable for development and implementation of the EI system should be established at the outset. The planning committee should include representatives from all ranks of the department. (This important issue is discussed in more detail later.) Key Issue: Create a structured planning process. The planning process should involve a series of regular meetings (weekly, bi-weekly, and monthly) and these meetings should continue with interruptions only for special circumstances. The development and implementation of EI systems in some agencies took as long as two years. While in some cases the delay was the result of unavoidable issues related to computer hardware or software, in others it was because department leaders deliberately took the time to involve all stakeholders and resolve possible conflicts. Officials involved in those processes believe that taking the time was worth the effort. As one commander put it, "some important things would have been missed" if they had rushed the development too quickly. The most important things that would have been missed were (1) attention to certain issues that could have caused problems later on, and (2) the buy-in from officers that resulted from their having a chance to become familiar with the nature and purpose of the system. The importance of taking the time to involve stakeholders and resolve all issues in advance is documented in the COPS Office report, Mediating Citizen Complaints: A Guide for Police and Community Leaders (2002). STEP FOUR: Pinpoint Responsibility A commander with the rank of Deputy Chief or Captain should be appointed chair of the planning committee with responsibility for managing the process. Key Issue: Assign and hold one official responsible for the planning process and the final development of the program. STEP FIVE: Involve Officers From All Ranks Officers representing all ranks of the department should be appointed to the planning committee. Representation from all ranks serves two purposes: Firstly, the planning process benefits from the insights of staff members who have different experiences and perspectives on how an EI system should operate. Sergeants, who play a key role in a functioning EI system, will have insights into potential problems regarding how the system will be integrated into their normal duties. Direct involvement in the planning and the opportunity to raise important issues ensures buy-in from all ranks in the department. Secondly, the members of the committee become an important channel of communication to other staff members, allowing them to explain the concept and reduce the potential harm of rumors and misinformation. Key Issue: Involve officers from all ranks directly in the planning process. The planning process should represent the full range of rank, experience, race, ethnicity, and gender in the agency. Representatives of the collective bargaining unit for rank and file officers should also be included in the planning process. It is important to avoid creating a system that might conflict with certain clauses in the collective bargaining contract. If the EI system covers non-sworn employees in the agency, they should also be included in the planning process. And if they are represented by a collective bargaining unit, officials from that organization should be involved. Key Issue: Involve representatives from the collective bargaining unit(s). STEP SIX: Utilize Outside Experts There is now a substantial body of experience with EI systems in the law enforcement community. No agency needs to reinvent the wheel at this point. This report is designed to serve as a guide to the issues related to developing an EI system. Law enforcement officials in departments with operating EI systems currently in place are particularly useful as resources. They bring their practical experience with EI systems and as peers have substantial credibility with officers in the department. Key Issue: Don't try to reinvent the wheel. Utilize existing expertise from other law enforcement agencies or other experts. Issues such as the selection of the thresholds for identifying officers and choice of software are areas where an outside consultant can be particularly useful. STEP SEVEN: Assess the Department's Computer Infrastructure Based on the information learned from other departments, what is the current state of the department's computer infrastructure? Is it capable of supporting the more sophisticated systems that exist in other departments? For example, some systems allow each first- line supervisor to log on to the system each day before roll call. Would that be possible in your department? If the department's computer system would not support the most sophisticated system, what alternatives are possible? Some of the best systems currently exist in departments with 2,000 and 8,000 sworn officers. Does a medium-sized department need as complex an EI system? What kind of system would be appropriate–and feasible–for a department of 300 to 700 sworn officers? Key Issue: Begin addressing computer infrastructure needs at the very outset. Plan a system that is both appropriate in terms of accountability and feasible for the department. STEP EIGHT: Keep the Staff Informed Members of the agency should be kept informed of the development of the EI system as it develops. Involving officers from all ranks on the planning committee helps address this issue, but other steps can be taken. In one agency a series of separate meetings were held with the captains, lieutenants, and sergeants. Keeping all members informed through regular presentations helps to overcome misunderstandings that might impede the efficient implementation of the system. In the PERF survey (See Chapter Three), failures to communicate with members of the agency were the major problem reported by managers experienced with EI systems. Key Issue:To avoid problems arising from misunderstanding, provide staff at all ranks with detailed information about the program. STEP NINE: Keep External Stakeholders Informed An EI system is an important instrument for strengthening accountability within a department. Equally important is the task of building the perception of accountability among important leaders and groups outside the department. These groups include the responsible elected officials (mayors, city mangers, city council or county board members, community groups, etc.). Advising the public about the commitment to creating an EI system and keeping them informed about its progress can help develop positive relations with the community. Key Issue: Build community and political support by keeping external stakeholders informed about the development of the EI system. STEP TEN: Identify and Address Specific Issues A number of specific issues need to be addressed in the planning process: (1) Performance Indicators The planning process should include extensive discussions of the performance indicators to be used by the EI system. This process will facilitate understanding of the nature of an EI system as a comprehensive personnel assessment system and not just a process for identifying officers with performance problems. Also, the process will highlight deficiencies in the agency's existing data systems that need to be corrected if the EI system is to function properly. As Chapter One explained, there is no consensus of opinion about the exact number of indicators that should be used (except that relying on only one or two is inappropriate). Existing systems use as few as eight and as many as 24. Planning discussions about the number of indicators should take into account the specific needs of the department (See Step One, above) and the capacity of the department to support a large versus a small system. To assist in planning efforts, examples of performance indicators from several departments are presented in the Appendix to this report. (2)Thresholds for Identifying Officers The planning process should involve extensive discussions about what thresholds to set for identifying officers needing intervention. Although this is an extremely important issue, there are currently no nationally recognized best practices regarding thresholds. A variety of different systems are being used (See Chapter One). This is one area where the use of outside experts would be particularly valuable so that the planning committee can learn from the practical experience of other agencies. The discussion of thresholds is an opportunity for members of the planning committee to think through current performance standards within the agency and to reach a consensus about appropriate standards. As part of its development process, the Phoenix Police Department tested proposed thresholds to get a sense of how they would work in practice. This involved loading past performance data into the system and then applying the thresholds retrospectively. Tests of this sort can help to ensure that thresholds do not identify an unmanageable number of officers or fail to identify potential problems in certain areas of police activity. To assist in planning efforts, examples of thresholds used in other departments are presented in the Appendix to this chapter. Key Issue:Think through routine operations and identify key performance indicators. One size does not necessarily fit all. Indicators used by other agencies may not be relevant or appropriate. Some other indicators may be appropriate for your agency. (3) Define the Format of Interventions The format of the interventions needs to be discussed. There is presently no national consensus regarding the best format. Some agencies rely on immediate supervisors to counsel subject officers. Others have meetings or counseling sessions with two or more supervisors present. Still others conduct classes for groups of officers selected for intervention. Discussions of alternative approaches will help officials better understand the purpose of the interventions and feel comfortable with the format ultimately selected. (4) Coordinate the EI System With the Disciplinary Process The EI system is designed to be an alternative to the normal disciplinary process. Once an EI system is operational the two systems will function along separate but often parallel tracks. It is important to clearly define the boundary between the two systems. For this reason, it is advisable for representatives of the internal affairs or professional standards unit to be included on the planning committee. Key Issue: Be sure to define clearly the relationship between the EI system and the existing disciplinary process. (5) Develop a List of Corrective Actions Officers conducting intervention sessions should have a number of specific recommendations for programs or steps that subject officers can or should take. These programs or steps might include retraining on specific police actions or a recommendation that the officer seek family counseling. The planning committee should develop this list of programs or steps. To assist in planning, examples of alternative corrective actions used by some departments are presented in the Appendix to this report Key Issue: Provide staff members who will be conducting the counseling sessions with practical and relevant alternatives that they can recommend to employees. (6) Decide What Kinds of Documentation Will Be Required For purposes of internal accountability, it is important to document the various steps in the EI system process. What kinds of reports about officers selected for intervention will be required? Will they be quarterly, semi-annual, annual? What will be included in these reports? How much detail will be included about each officer's performance? How will interventions be documented? Who will be responsible for reports about intervention sessions? How much detail will be included in these reports? What kind of post-intervention documentation will be required? For how long will a subject officer be monitored? Who will be responsible for that monitoring? What kind of evidence will be sufficient to remove an officer from the EI system? (7) Define the Day-to-Day Responsibilities of First-Line Supervisors What will be the day-to-day responsibilities of first-line supervisors? The Pittsburgh PARS requires sergeants to log on to the system each day prior to roll call. They are expected to review officer's performance records and identify any potentially troublesome patterns and take some appropriate action even before any officer exceeds a formal threshold. The purpose of this approach is to integrate the EI system into the normal supervisory process and to create a culture of accountability throughout the department. (8) Decide on Notification Procedures Some EI systems have the technical capacity to notify anyone from the first-line supervisor to the chief executive about any incident involving a particular officer. For example, if Officer X has been identified as having serious performance problems, the EI system can be programmed to send selected officials an email notification about selected performance indicators involving that officer (e.g., any use-of-force report, citizen complaint, etc.). Is this kind of process appropriate for the department? Does the department have an internal email system capable of supporting such a process? STEP ELEVEN: Take the Time to Get It Right Make sure all the issues are resolved before finalizing an EI system. Proceeding slowly but deliberately allows stakeholders an opportunity to gain an understanding of the EI system and to become comfortable with it. Some agencies developed their EI systems quickly then imposed on the department without warning. In these cases, the response has usually been suspicion, hostility, and resistance from officers at many ranks. Key Issue: EI Systems are very complex. It is better to take the time to get it right than to rush and create a system with unresolved issues. STEP TWELVE: Provide Staff Training All officers need to be trained in the operation of the system before it becomes operational. One of the problems identified in the PERF survey (See Chapter Three) was the failure of some departments to adequately explain the responsibilities of various officials in the EI system and to provide the proper training. Rank and file officers should be fully informed about the nature of the system and how it will operate in practice. Sergeants and any others who will be involved in interventions need to be trained in how to conduct intervention sessions. Lieutenants, captains, and other higher-ranking officers need to be trained regarding their responsibilities in the EI system. STEP THIRTEEN: Choose Software The planning committee should begin exploring software at the earliest moment. There are several commercial software packages on the market. A number of departments have developed their own software. The decision regarding software is an issue where the use of outside experts can be particularly valuable. STEP FOURTEEN: Finalize the Plan The last stage of the planning process should involve a final meeting of the planning committee to ensure that all issues have been addressed to the satisfaction of all members and to approve the final plan. The committee should review the final draft of the General Order and direct that it be sent to the chief executive for sign-off and official implementation of the EI system. STEP FIFTEEN: Implement the EI System Implement the Early Identification system. Chapter Six Policy and Research Needs for Early Intervention Systems I. Current Issues Regarding EI Systems . . . . . . . . . . . . . . . . . . . . . .109 II. The Future: An Action Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . .111 Policy and Research Needs for Early Intervention Systems Early Intervention (EI) systems are presently still evolving. After a number of years of development, law enforcement agencies now have considerable experience with developing and operating EI systems. Even the best systems, however, continue to evolve as the managers responsible for them attempt to improve them. While there is now a rough consensus of opinion within the field regarding the basic elements of an effective EI system, many important questions remain unanswered. The purpose of this chapter is to discuss the most important questions regarding EI systems and to suggest the appropriate policy and research needs. I. Current Issues Regarding EI Systems (1) What are the appropriate performance indicators? While there is a strong consensus of opinion that an EI system should not rely on a single performance indicator (such as citizen complaints) there is no consensus regarding the exact number that should be used. There is, however, a strong consensus of opinion that certain indicators, such as use-of-force reports, citizen complaints, etc., should always be among those indicators chosen. The question of the appropriate number of indicators raises a more fundamental issue of the size and scope of an EI system. Large and complex EI systems have the capacity for comprehensive personnel assessment. They can identify top-performing officers as well as poor performers. Such systems, however, are not only expensive but pose a number of management challenges in terms of computer infrastructure, data entry, and ongoing management oversight. A system that is not managed properly will not be effective, no matter how sophisticated it is in theory. Smaller EI systems that focus on officers with performance problems and utilize fewer performance indicators cannot do as much in terms of personnel assessment, but are far easier to create and manage. Further research and evaluation is needed with regard to the mot effective size and scope of an EI system, and the related question regarding the proper number of performance indicators. (2) What are the appropriate thresholds for identifying officers for intervention? Several different systems of thresholds for identifying officer for possible intervention are currently in use. They include department-level thresholds, peer-officer thresholds, and ratios among different indicators (See Chapter Two). It is not clear at this point which formula is the best. Further research and evaluation is needed regarding the effectiveness of different systems of thresholds for identifying officers for possible intervention. (3) What is the best procedure for ensuring that supervisors responsible for conducting intervention sessions perform this task in a manner consistent with departmental goals? The intervention component of an EI system is the crucial part of the attempt to improve an officer's performance. An EI system involves a shift in the role of supervisors and the development of skills that many supervisors may not have or have been trained for. Additional research and evaluation is needed of different approaches for training supervisors for their new roles in an EI system. (4) What are the most effective kinds of intervention? It is likely that some forms of intervention are more effective than others for certain kinds of performance problems. Little attention has been given to this issue. Further research and evaluation should be given to developing more effective ways of tailoring specific interventions to particular officer performance problems. (5) What is the best size and scope of EI systems for small or medium-sized law enforcement agencies? Almost all of the information about EI systems available at present relates to large law enforcement agencies. Additional program development is needed with regard to EI systems appropriate for small and medium-sized law enforcement agencies. II. The Future: An Action Agenda Policy Development EI systems are developing all over America. Police managers associated with even the most sophisticated systems in the country continue to encounter new issues that need to be resolved. There is at present a substantial body of experience with EI systems and a cadre of police managers who have thought about the issues they face. The law enforcement profession should draw upon this experience and sponsor a series of activities designed to further the development of EI systems. Specifically, we recommend that: (1) Law enforcement professional associations sponsor a regular series of workshops and conferences specifically dedicated to the development and management of Early Intervention systems. Experience to date indicates that EI systems are extremely complex administrative tools and, consequently, it is not anticipated that the major issues will be resolved quickly. Consequently, we recommend that the law enforcement professions establish a regular series of workshops and conferences specifically dedicated to issues surrounding EI systems.143 (2) Law Enforcement professional associations undertake the development of detailed recommended Standards for Early Intervention Systems. One purpose of the workshops and conferences should be to begin the development of recommended Standards for Early Intervention Systems. These would be very detailed and provide guidance to the profession on such issues as performance indicators, thresholds, and so on. It is expected that these standards would evolve over time, as a better understanding of EI systems develops. (3) In developing the Standards, law enforcement professional associations need to specifically address the unresolved issues discussed earlier in this chapter. Research Further research is need on the nature and impact of EI systems. Consequently, we recommend that the federal government and private funding agencies: (4) Fund research on the effectiveness of EI systems in reducing problematic performance among officers subject to intervention. (5) Fund research on the relative effectiveness of different EI systems with regard to such issues as their performance indicators and intervention strategies. (6) Fund research on the relationship of EI systems to other law enforcement accountability mechanisms. The Intersection of Research and Policy All too often, academic research is not effectively translated into operational policy. To overcome this problem, we recommend that the federal government and private funding agencies: (7) Fund workshops bringing together police officials responsible for policy development and academic researchers. Appendices Appendix A .................................................117 Appendix B .................................................120 Appendix C .................................................124 Appendix D .................................................126 Appendix E..................................................130 Appendix F..................................................137 Reading List ...............................................148 Endnotes ...................................................150 Appendices Appendix A Consent Decree, United States v. Los Angeles (2001). The entire consent decree can be found on the U.S. Department of Justice website: www.usdoj.gov/crt/split. 41. TEAMS II shall contain information on the following matters: a. all non-lethal uses of force that are required to be reported in LAPD "use of force" reports or otherwise are the subject of an administrative investigation by the Department; b. all instances in which a police canine bites a member of the public; c. all officer-involved shootings and firearms discharges, both on-duty and off-duty (excluding training or target range shootings, authorized ballistic testing, legal sport shooting events, or those incidents that occur off-duty in connection with the recreational use of firearms, in each case, where no person is hit by the discharge); d. all other lethal uses of force; e. all other injuries and deaths that are reviewed by the LAPD Use of Force Review Board (or otherwise are the subject of an administrative investigation); f. all vehicle pursuits and traffic collisions; g. all Complaint Form 1.28 investigations; h. with respect to the foregoing clauses (a) through (g), the results of adjudication of all investigations (whether criminal or administrative) and discipline imposed or non-disciplinary action taken; i. all written compliments received by the LAPD about officer performance; j. all commendations and awards; k. all criminal arrests and investigations known to LAPD of, and all charges against, LAPD employees; l. all civil or administrative claims filed with and all lawsuits served upon the City or its officers, or agents, in each case resulting from LAPD operations, and all lawsuits served on an officer of the LAPD resulting from LAPD operations and known by the City, the Department, or the City Attorney's Office; m. all civil lawsuits against LAPD officers which are required to be reported to the LAPD pursuant to paragraph 77; n. all arrest reports, crime reports, and citations made by officers, and all motor vehicle stops and pedestrian stops that are required to be documented in the manner specified in paragraphs 104 and 105; o. assignment and rank history, and information from performance evaluations for each officer; p. training history and any failure of an officer to meet weapons qualification requirements; and q. all management and supervisory actions taken pursuant to a review of TEAMS II information, include non-disciplinary actions. TEAMS II further shall include, for the incidents included in the database, appropriate additional information about involved officers (e.g., name and serial number), and appropriate information about the involved members of the public (including demographic information such as race, ethnicity, or national origin). Additional information on officers involved in incidents (e.g., work assignment, officer partner, field supervisor, and shift at the time of the incident) shall be determinable from TEAMS II. Appendix B Los Angeles Sheriffs Department, Personnel Performance Index (PPI), SHERIFF'S BULLETIN #443 June 3,1997 Performance Review The purpose of this bulletin is to explain the performance review process, Background Over the past years, I have consistently expressed my belief that the Department's most important resource is our employees. Performance review and follow-up is one of the most significant ways that the Department demonstrates its commitment to supporting its employees. The tasks of reviewing, assessing, and guiding subordinate performance are not new ones. They have always been, and will continue to be, supervisory responsibilities. However, the structure and procedures that comprise the career performance review process that is the subject of this bulletin are new. having been implemented in the early part of 1996. At that time each division contribute a one supervisor to create a performance review staff. In addition, the eight Commanders who were serving on the Executive Risk Management Committee were designated to serve as members of the Performance Review Committee. Identification of Reviewees There is more than one method of determining which personnel are to be selected for review. One method starts with an examination of summary performance information in the Personal Performance Index (PPI). The PPI facilitates the identification of employees who have been involved in disproportionate numbers of (and/or very serious) risk incidents. Another method of identifying candidates to be reviewed involves supervisors' and managers' recommendations based on long-term observations and knowledge about their subordinates' work. Performance Review Report Upon designation of a candidate, the performance review staff collects virtually all the performance documentation available from the beginning of the employee's career. The material assembled includes photocopies of performance evaluations, commendations, notifications of discipline, and time records, as well as the detailed documentation on the risk incidents of interest –shootings, uses of force, public complaints, administrative investigations, preventable traffic collisions, civil claims, and lawsuits. The staff member then compiles a profile report recounting the individuals career performance history as reflected in the documentation. The report and the accompanying documentation form the basis for the review process as conducted by the Performance Review Committee Commanders. Performance Review Committee The Commanders who comprise the Performance Review Committee meet in rotating groups of three members approximately twice per month as the profile repor1s are completed. The group is chaired by the Commander assigned to the Office of Administrative Services, who is the only Commander who does not rotate. The meetings are also attended by the Depar1ment Training Administrator, representatives from Employee Relations! Advocacy Services, Risk Management Bureau suppor1 staff, and the employee's Unit Commander. The committee members are furnished copies of the profile report prior to the scheduled date of the review. At the meeting, the committee evaluates the contents of the profile report and the compiled documentation and decides whether it is necessary to develop a plan for guiding the employee's future performance. Not every employee who is selected for a performance review is deemed to require a performance guidance plan. The committee carefully assesses each employee's situation and mayor may not conclude that a reviewed individual presents a performance risk to himself/herself or to the Department. Formation of Performance Plan In cases where it is determined that an employee performance guidance plan is appropriate, the committee Commanders, along with the Unit Commander, confer to develop a plan. Elements of plans may range from discussions with the employee. more training, or temporary work reassignments up to, in extreme cases, fitness for duty evaluations. One feature that is an element of each plan is the Unit Commander's in-person discussion with the employee. explaining the concerns about his/her performance as it impacts both the employee and the Department. Employees are entitled to review the profile report and the documentation upon which the committee's decision was based. Review of Management Personnel And Systems Reviewing an individual employee's performance is only one objective of this process. Another goal is to assess the effectiveness of the management/supervisory systems and personnel at the current and previous units of assignment. In some cases, it has been determined that a supervisor's or manager's responsibility to assist an employee in improving performance has not been fulfilled. In these cases, the review committee Commanders contact the concerned supervisors and/or managers to counsel them regarding these supervisorial issues and reinforce the message that genuine concern for subordinate welfare is demonstrated by taking the time and energy to be honest, supportive, and helpful. Results of Performance Review As of April 1997, 144 employees have been the recipients of performance reviews–fewer than two percent of the Department's sworn personnel. Only two people have been recommended for evaluations of fitness for duty. Conclusion The performance review process is not disciplinary in nature. It was developed and implemented to provide a fair and responsible system for improving or enhancing the employee's performance. Emphasis is placed on the concern for correcting the employee's work performance, as opposed to administering discipline. The process benefits a variety of individuals, including the reviewee as well as his or her current and previous supervisors and managers. It integrates concern for individual employee interest with good stewardship of the Department's collective welfare and the employee's future well being. Appendix C Miami-Dade Police Department, Employee Identification System (EIS) (1) systematic review of complaints and use of Force incidents by sworn/nonsworn personnel 2) Identify stress or performance related problems c. Report. composition 1) Quarterly/Annual Report criteria a) Two or more personnel complaints or: b) Involvement in three or more Use of Force incidents c) Annual Report – Employee identified in two or more quarters; four or more complaints, seven or more Supervisor's Use of Force Report to Control, or two or more shooting incidents in twelve months 2) Report format a) Brief profile of complaints and Use of Force Incidents: (1) Complaints profile will include employee's name, badge number, complainant's name, nature of complaint, and disposition. (2) Use of Force profiles will include employee's name, badge number, subject's name, date of incident, nature of incident, extent of injury. d. Report Disposition 1) Report disseminated from Director's Office to employee's chain of command 2) Summary of supervisory review will be submitted to Internal Review Bureau within 30 days 3) Action alternatives a) Referral to Psychological services Program for counseling or referral assistance b) Participation in stress Abatement Program for training assistance (40 hours) c) Corrective action by supervisory personnel; counseling, discipline, termination d) Assessment that no problem exists, terminating further action Appendix D Risk Management System, Cincinnati Police Department Memorandum of Understanding, U.S. Department of Justice and the City of Cincinnati (2002) VII. Management and Supervision A. Risk Management System 57. The CPD will enhance and expand its risk management system to include a new computerized relational database for maintaining, integrating, and retrieving data necessary for supervision and management of the CPD. Priority will be given to the CPD obtaining any established program and system. The CPD will regularly use this data to promote civil rights and best police practices; to manage risk and liability; and to evaluate the performance of CPD officers across all ranks, units and shifts. 58. The new risk management system will collect and record the following information: (a.) all uses of force; (b.)canine bite ratios; (c.) the number of canisters of chemical spray used by officers; (d.)all injuries to prisoners; (e.) all instances in which force is used and a subject is charged with "resisting arrest," "assault on a police officer," "disorderly conduct," or "obstruction of official business;" (f.) all critical firearm discharges, both on-duty and off-duty; (g.)all complaints (and their dispositions); (h.)all criminal proceedings initiated, as well as all civil or administrative claims filed with, and all civil lawsuits served upon, the City, or its officers, or agents, resulting from CPD operations or the actions of CPD personnel; (i.) all vehicle pursuits; (j.) all incidents involving the pointing of a firearm (if any such reporting requirement is imposed by paragraph 25); and (k.)all disciplinary action taken against officers. 59. The new risk management system will include, for the incidents included in the database, appropriate identifying information for each involved officer (e.g., name, badge number, shift and supervisor) and civilian (e.g., race, ethnicity or national origin). 61. The CPD will, within 90 days, prepare for the review and approval of DOJ, and thereafter implement, a protocol for using the risk management system. The City will submit for the review and approval of DOJ all proposed modifications to the protocol prior to implementing such modifications. 62. The protocol for using the risk management system will include the following provisions and elements: (a.) The protocol is comprised of the following components: data storage, data retrieval, reporting, data analysis, pattern identification, supervisory assessment, supervisory intervention, documentation and audit. (b.)The protocol will require the automated system to analyze the data according to the following criteria: i) number of incidents for each data category by individual officer and by all officers in a unit; ii) average level of activity for each data category by individual officer and by all officers in a unit; and iii) identification of patterns of activity for each data category by individual officer and by all officers in a unit. (c.) The protocol will require the system to generate reports on a monthly basis describing the data and data analysis and identifying individual and unit patterns. (d.)The protocol will require that CPD commanders, managers, and supervisors will review, on a regular basis but not less than quarterly, system reports, and will evaluate individual officer, supervisor, and unit activity. (e.) The protocol will require that CPD commanders, managers, and supervisors initiate intervention for individual officers, supervisors and for units based on appropriate activity and pattern assessment of the information contained in the risk management system. (f.) The protocol will require that intervention options include discussion by commanders, managers, supervisors, and officers; counseling; training; and supervised, monitored, and documented action plans and strategies designed to modify activity. All interventions will be documented in writing and entered into the automated system (appropriate intervention options will be employed based on the evaluation described in subsection (e) above). (g.)The protocol will specify that actions taken as a result of information from the risk management system be based on all relevant and appropriate information, including the nature of the officer's assignment, crime trends and crime problems, and not solely on the number or percentages of incidents in any category of information recorded in the risk management system. (h.)The protocol will require that the system's data be accessible to CPD commanders, managers, and supervisors. Commanders, managers, and supervisors will promptly review records of all officers recently transferred to their sections and units. (i.) The protocol will require that CPD commanders, managers, and supervisors be evaluated on their ability to use the risk management system to enhance effectiveness and reduce risk. (j.) The protocol will require that the system be managed and administered by the Inspections Unit of the CPD. The Inspections Unit of the CPD will conduct quarterly audits of the system to ensure action is taken according to the process described above. (k.)The protocol will require regular reviews, at no less than quarterly intervals, by appropriate managers of all relevant risk management system information to evaluate officer performance city wide, and to evaluate and make appropriate comparisons regarding the performance of all CPD units in order to identify any significant patterns or series of incidents. Appendix E Consent Decree, United States v. State of New Jersey (1999). Supervisory Measures to Promote Civil Rights Integrity Management Awareness Program 40. The State shall develop and implement computerized systems for maintaining and retrieving information necessary for the supervision and management of the State Police to promote professionalism and civil rights integrity, to identify and modify potentially problematic behavior, and to promote best practices (hereinafter, the "Management Awareness Program" or "MAP"). 41. The MAP shall consist of the following information: a. all items of information in connection with all motor vehicle stops that are required to be recorded in a written report, form, or log, or reported to the communications center, pursuant to 29 and the protocols listed in 29of this Decree, except that duplicate information need not be entered, and information as to whether the incident was recorded with MVR equipment need not be entered if all patrol cars are equipped with MVR unless a patrol car was equipped with MVR equipment that was not functioning; b. information on civilian compliments and other indicia of positive performance; information on misconduct investigations reports on use of force associated with motor vehicle stops; on-duty and off-duty criminal arrests and criminal charges; civil suits involving alleged misconduct by state troopers while on duty; civil suits in which a trooper is named as a party involving off-duty conduct that alleges racial bias, physical violence or threats of violence; and c. implementation of interventions; and training information including the name of the course, date started, date completed and training location for each member receiving training. 42. All information in MAP on substantiated misconduct investigations, civilian compliments, and other indicia of positive performance which can be attributed to a specific trooper shall be made available to that trooper on an annual basis upon written request. Nothing in this paragraph shall be construed as granting that trooper access to confidential documents other than those identified in this paragraph, or to any information which cannot be attributed to the trooper requesting the information. 43. Regarding the motor vehicle stop information identified in 29 (a) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, and 19) and recorded in accordance with the protocols identified in 29(a), the MAP shall have the capability to search and retrieve numerical counts and percentages for any combination of the above-referenced information and to run reports for different time periods (e.g., monthly, quarterly, annually) and for individual troopers, squads, and stations. Regarding the motor vehicle stop information identified in 29(a)(5A, 8A, 12A, 13A, 14A, 15A, and 17A) and recorded in accordance with the protocols identified in 29(a), it will be sufficient that the MAP shall have the capability to access (through cross-referenced paper documents or other method) this descriptive information entered on specific incidents and matters. Regarding the information identified in 41(b and c), to the extent technologically feasible, the MAP shall be developed to have the capability to search and retrieve numerical counts and percentages for any combination of the information and to run reports for different time periods and for individual troopers, squads or stations. To the extent that the MAP shall require textual or narrative descriptions of misconduct allegations or other information identified in 41(b and c), it will be sufficient that the MAP only have the capability to retrieve this descriptive information. 44. Where information about a single incident is included within the MAP from more than one document the State shall use a common control number or other means to link the information from different sources so that the user can cross-reference the information and perform analyses. 45. The State shall ensure that information is included within the MAP in an accurate and timely fashion and is maintained in a secure manner. 46. Within one hundred and eighty (180) days following entry of this Decree, the State shall develop a plan for designing and implementing the MAP including the use of the MAP, a timetable for implementation, and a specification of the information contained in State records pre-dating the implementation of the MAP that can reasonably be incorporated in the MAP. Prior to effectuating the implementation plan, the plan shall be approved by the United States and the Independent Monitor. Within 180 days following the entry of this Decree, the State shall begin conducting the supervisory and management reviews required by 48-53. Supervisory and Management Reviews 47. Consistent with the requirements of 48-53 infra, the State shall develop a protocol specifying the manner in which supervisory and management reviews of individual state troopers, and State Police units and sub-units (e.g., troops, stations, and squads), shall be conducted, and the frequency of such reviews. Prior to implementation, the protocol shall be approved by the United States and the Independent Monitor. 48. At least quarterly, State Police supervisors shall conduct reviews and analyses of data obtained from the MAP and other appropriate sources to ensure that individual troopers and State Police units and subunits are performing their duties in accord with the provisions of this Decree and associated protocols. 49. To the extent reflected in 43, reports of MAP data shall regularly be prepared regarding individual troopers, stations and squads, for use in reviews as appropriate. The reports shall include the following information: a. the number of motor vehicle stops, by race/ethnicity, reason for the stop (i.e., moving violation, non moving violation, other), road, squad, and trooper station; and the number of enforcement actions and procedures taken in connection with or during the course of a motor vehicle stop, by race/ethnicity, reason for the stop (i.e., moving violation, non- moving violation, other), road, squad and trooper station; b. data (including racial/ethnic data) on complaints, misconduct investigations (for each type of investigation, as delineated in 73), discipline, intervention, and uses of force associated with motor vehicle stops. 50. To the extent reflected in 43,analyses of MAP data concerning motor vehicle stops shall include a comparison of racial/ethnic percentages of motor vehicle stops (by reason for the stop (i.e., moving violation, non moving violation, other)) and racial/ethnic percentages of enforcement actions and procedures taken in connection with or during the course of such stops, with a benchmark racial/ethnic percentage if available (see 54-55); a comparison of racial/ethnic percentages for such stops with the racial/ethnic percentages for enforcement actions taken in connection with or the during the course of such stops; a comparison of racial/ethnic percentages for consent searches of vehicles, and requests for consent to search vehicles, with "find" rates by race/ethnicity for motor vehicle consent searches; a comparison of racial/ethnic percentages for non-consensual searches of motor vehicles with "find" rates by race/ethnicity for motor vehicle non-consensual searches; evaluations of trends and differences over time; and evaluations of trends and differences between troopers, units, and subunits. 51. To the extent reflected in 43,analyses of other data generated by the MAP shall include evaluations of trends and differences over time and evaluations of trends and differences between troopers, units, and subunits. 52. Each supervisor shall, consistent with his or her authority, implement any appropriate changes or remedial measures regarding traffic enforcement criteria, training, and enforcement practices for particular units or subunits or implement any appropriate intervention for particular troopers; conduct any necessary additional assessment or investigation regarding particular units or subunits or particular troopers; and/or make any appropriate recommendations. 53. A supervisory review shall be conducted regarding any state trooper who within a period of two years, is the subject of three misconduct investigations of any kind initiated pursuant to 73. Where appropriate, the review may result in intervention being taken. In the event the supervisory review results in intervention, the supervisor shall document the nature, frequency, and duration of the intervention. 54. To assist in evaluating data reported from the MAP concerning State Police law enforcement on the New Jersey Turnpike, the State shall develop (for purposes of implementing this Decree) a protocol for conducting a survey of a sample of persons and vehicles traveling on the New Jersey Turnpike to determine the racial/ethnic percentage of drivers on the Turnpike. As appropriate, the survey may identify different benchmark figures for different portions of the Turnpike. Prior to implementation, the protocol shall be approved by the Independent Monitor and the United States. The protocol shall be developed and implemented using a consultant jointly selected by the parties. The survey shall be completed within one hundred fifty (150) days of the entry of this Decree. Both the United States and the State agree that the utility and fairness of the MAP described in this Consent Decree will depend to some degree on the development of accurate and reliable benchmarks that account for all appropriate variables and factors. 55. For purposes of implementing this Decree, the State may in its discretion conduct a survey of a sample of persons and vehicles traveling on any other limited access highway in New Jersey to determine the racial/ethnic percentage of drivers on that road. In deciding whether to conduct such a survey, the State shall give due consideration to any recommendations made by the United States and the Independent Monitor. The survey shall be conducted pursuant to a protocol developed by the State and approved by the Independent Monitor and the United States. The protocol shall be developed and implemented using a consultant jointly selected by the parties. 56. The parties agree that for purposes of developing appropriate benchmark racial/ethnic percentages, it will be sufficient for the surveys described by 54 and 55 to determine the racial/ethnic percentages of persons and vehicles traveling on the subject roadway. The provisions of 54 and 55 shall not apply to any traffic survey conducted by the State for purposes of advocating a position in any criminal case or any civil litigation other than in the instant case. Appendices Appendix F EI Systems: An Overview • Data-based management information system. • Capacity for identifying and correcting performance problems. • Recommended police accountability "best practice" • Separate from the formal disciplinary system • Consistent with the goals of Ccommunity Ppolicing • Consistent with the process of Pproblem-Ooriented Ppolicing • Rely on systematic and timely data • Consistent with the process of COMPSTAT • Careful planning is needed for development and implementation Office of Community Oriented Policing Services U.S. Department of Justice Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha COPS COMMUNITY OORIENTED PPOLICING SSERVICES U.S. DDEPARTMENT OOF JJUSTICE Goals and Impacts of EI Systems The Individual Officer • Improved Performance • Higher Standards of Accountability Supervisors • Improved Supervisory Practices • Higher Standards of Accountability as Supervisors The Department • Higher standards of accountability • Reduction in litigation • Improved Community Relations Office of Community Oriented Policing Services U.S. Department of Justice Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha COPS COMMUNITY OORIENTED PPOLICING SSERVICES U.S. DDEPARTMENT OOF JJUSTICE Components of an EI System • Performance Indicators • Identification and Selection • Intervention • Post-Intervention Monitoring Office of Community Oriented Policing Services U.S. Department of Justice Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha COPS COMMUNITY OORIENTED PPOLICING SSERVICES U.S. DDEPARTMENT OOF JJUSTICE Performance Indicators Comprehensive Personnel Assessment Model 1. All non-lethal uses of force Citizen injuries requiring medical attention Disciplinary outcomes 2. All officer-involved shooting incidents On-duty/off-duty discharges, citizen injuries, citizen fatalities Disciplinary outcomes 3. All officer-involved vehicular pursuits Accidents Injuries Fatalities Disciplinary outcomes 4. All citizen-initiated complaints Use of force Racial bias Investigative outcomes 5. All citizen-initiated commendations or compliments 6. All departmental commendations and awards 7. Criminal arrests and investigations of subject officer 8. Civil suits or an administrative claim in which subject officer is named Nature of claims Case outcomes Office of Community Oriented Policing Services U.S. Department of Justice Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha COPS COMMUNITY OORIENTED PPOLICING SSERVICES U.S. DDEPARTMENT OOF JJUSTICE Performance Indicators (continued) 9. All arrest reports, crime reports, and citations made by officers 10. All motor vehicle stops and pedestrian stops made by subject officer Race, ethnicity, gender, age of drivers stopped 11. Performance evaluations for each officer 12. Training history Any failure of an officer to meet weapons qualification requirements 13. All management and supervisory actions taken pursuant to a review of TEAMS II information, including non-disciplinary actions 14. Sick leave or family leave record 15. Canine unit deployment involving subject officer Deployments Bites Citizen injuries Complaints or civil suits arising from deployments 16. Failure to appear in court Office of Community Oriented Policing Services U.S. Department of Justice Early Intervention Systems for Law Enforcement Agencies: A Planning and Management Guide Prepared by Samuel Walker, PhD Department of Criminal Justice University of Nebraska at Omaha COPS COMMUNITY OORIENTED PPOLICING SSERVICES U.S. DDEPARTMENT OOF JJUSTICE General Assessment of EI System Image of a bar graph showing the general assessment of EI System Impact of EI System on Quality of Police Service Image of a bar graph showing the impact of the EI System on Quality Reading List Christopher Commission, Report of the Independent Commission on the Los Angeles Police Department. Los Angeles: Christopher Commission, 1991. Commission on Accreditation for Law Enforcement Agencies. Standards for Law Enforcement Agencies. 4th ed. Fairfax, VA: CALEA, 1999. www.calea.org Robert C. Davis, et. al. Turning Necessity into Virtue: Pittsburgh's Experience with a Federal Consent Decree. New York: Vera Institute of Justice, 2002. Available at: www.vera.org. International Association of Chiefs of Police, Building Integrity and Reducing Drug Corruption in Police Departments. Gaithersburg: IACP, 1989. Kolts, James G., and Staff. The Los Angeles County Sheriff's Department. Los Angeles: Los Angels County, 1992. Los Angeles Police Commission. Report of the Rampart Independent Review Panel. Los Angeles: Los Angeles Police Commission, 2000. Special Counsel Merrick J. Bobb, 16th Semiannual Report. Los Angeles: Special Counsel, 2003. Available at: www.parc.info. Walker, Samuel. "The New Paradigm of Police Accountability," St. Louis University Public Law Review, XII (March 2003): 3–52. Walker, Samuel, Geoffrey P. Alpert and Dennis Kenney. Early Warning Systems: Responding to the Problem Police Officer. Research in Brief. Washington, DC: National Institute of Justice, 2001. Available on the web at www.ncjrs.gov. NCJ 188565. Weisburd, David, Stephen Mastrofski, Ann Marie McNally, and Rosann Greenspann, Compstat and Organizational Change: Findings from a National Survey. Washington, DC: The Police Foundation, 2001. U.S. Commission on Civil Rights, Who is Guarding the Guardians? A Report on Police Practices Washington, DC: U.S. Civil Rights Commission, 1981. U.S. Department of Justice, Principles for Promoting Police Integrity. Washington: Department of Justice, 2001. Available on the web at www.ncjrs.gov, NCJ # 186189. Endnotes 1 Samuel Walker, Geoffrey P. Alpert, and Dennis Kenney. 2001. Early Warning Systems: Responding to the Problem Police Officer. Research in Brief. (Washington, DC: National Institute of Justice, 2001). Available on the web at www.ncjrs.gov. NCJ 188565. 2 International Association of Chiefs of Police, Building Integrity and Reducing Drug Corruption in Police Departments (Washington, DC: Government Printing Office, 1989), p. 80. 3 Frank Landy, Performance Appraisal in Police Departments. (Washington, DC: The Police Foundation, 1977). 4 Bob Stewart, Comments, Focus Group Session, Nebraska State Patrol, June 9, 2003. 5 Gary Stix, "Bad Apple Picker: Can A Neural Network Help Find Problem Cops?" Scientific American (December 1994): 44–45. Bernard Cohen and Jan M. Chaiken, Police Background Characteristics and Performance (Lexington, Mass., Lexington Books, 1973). 6 Herman Goldstein, Police Corruption (Washington, DC: The Police Foundation, 1975). 7 It is significant, for example, that a recent National Institute of Justice (NIJ) publication on developing programs to deal with law enforcement officer stress includes a section on "Selecting Target Groups" but contains no reference to specific performance indicators such as are commonly used in EI systems. Peter Finn and Julie Esselman Tomz, Developing a Law Enforcement Stress Program for Officers and Their Families (Washington: Government Printing Office, 1997), pp. 23–26. Available on the web at www.ncjrs.gov. NCJ163175. 8 David Weisburd, Stephen Mastrofski, Ann Marie McNally, and Rosann Greenspann, Compstat and Organizational Change: Findings from a National Survey (Washington, DC: The Police Foundation, 2001). Eli B. Silverman, NYPD Battles Crime: Innovative Strategies in Policing (Boston: Northeastern University Press, 1999). 9 Robert L. Mathis and John H. Jackson, Human Resource Management. Michael Poole and Malcolm Warner, The IEBM Handbook of Human Resource Management (London: International Thomson Business Press, 1998). 10 One large city on the West Coast the union demanded that the EI system be subject to collective bargaining or meet and confer. The issue remained unresolved at the time this report was written. Nonetheless, this remains the only known case of union opposition to date. 11 Walker, Alpert, and Kenney, Early Warning Systems: Responding to the Problem Police Officer. 12 U.S. Department of Justice, Principles for Promoting Police Integrity (Washington: Department of Justice, 2001). Available on the web at www.ncjrs.gov, NCJ # 186189. 13 Commission on Accreditation for Law Enforcement Agencies, Standard 35.1.15. "Personnel Early Warning System." (2001). Commission on Accreditation for Law Enforcement Agencies, Standards for Law Enforcement Agencies, 4th ed. (Fairfax, VA: CALEA, 1999). www.calea.org 14 International Association of Chiefs of Police, Building Integrity and Reducing Drug Corruption in Police Departments. 15 U. S. Commission on Civil Rights, Who is Guarding the Guardians? (Washington DC: The United States Commission on Civil Rights, 1981). 16 The various consent decrees and memoranda of understanding, along with other related documents, are available at www.usdoj.gov/crt/split. 17 United States v. State of New Jersey, Consent Decree (1999), Parag. 40. Available at www.usdoj.gov/crt/split. 18 United States v. Los Angeles, Consent Decree (2000), Section II. Available at www.usdoj.gov/crt/split. 19 Department of Justice, Principles for Promoting Police Integrity. 20 Robert C. Davis, Christopher Ortiz, Nicole J. Henderson, Joel Miller and Michelle K. Massie, Turning Necessity into Virtue: Pittsburgh's Experience with a Federal Consent Decree (New York: Vera Institute of Justice, 2002), p. 37. Available at www.vera.org. 21 Samuel Walker, "The New Paradigm of Police Accountability: The U.S. Justice Department 'Pattern or Practice' Suits in Context," St. Louis University Public Law Review, XXII (No. 1, 2003): 3–52. 22 Walker, Alpert, and Kenney, Early Warning Systems: Responding to the Problem Police Officer. 23 Early Intervention System State of the Art Conference, Report (Omaha, 2003). Available at www.policeaccountability.org. 24 Davis, et al., Turning Necessity into Virtue, pp. 44 (quote), 43. 25 United States v. City of Los Angeles, Consent Decree, Sec. II, Para. 47. 26 The Los Angeles and other consent decrees are available on the U.S. Department of Justice website: www.usdoj.gov/crt/split. 27 United States v. Cincinnati, 2002: Sec. VII. 62. i. 28 San Jose Independent Police Auditor, 2001 Year End Report (San Jose: Independent Police Auditor, 2002), pp. 48–49. 29 This program is described in Special Counsel Merrick J. Bobb, 16th Semiannual Report (Los Angeles: Special Counsel, 2003), pp. 79–80. 30 United States v. State of New Jersey. 31 Unpublished Remarks. Early Intervention State of the Art Conference (2003). 32 American Correctional Associatioin, Performance-Based Standards for Adult Community Residential Services, 4th ed. (Lanham, MD: American Correctional Association, 2000), Appendix B, p. 82. 33 Timothy N. Oettmeier and Mary Ann Wycoff, Personnel Performance Evaluations in the Community Policing Context (Washington, DC: The Police Foundation, 1997). 34 Geoffrey Alpert and Mark H. Moore, "Measuring Police Performance in the New Paradigm of Policing," Bureau of Justice Statistics, Performance Measures for the Criminal Justice System (Washington: Government Printing Office, 1993), pp. 109–140. 35 Ibid. 36 Herman Goldstein, "Improving Policing: A Problem- Oriented Approach," Crime and Delinquency, 25 (1979): 236–258. Tara O'Connor Shelley and Anne C. Grant, eds., Problem-Oriented Policing (Washington: Police Executive Research Forum, 1998). 37 Davis, et al., Turning Necessity into Virtue, pp. 45–47. 38 Carol Archbold, Innovations in Police Accountability: An Exploratory Study of Risk Management and Police Legal Advising. (March 2002). Available at www.policeaccountability.org. 39 National Advisory Commission on Civil Disorders, Report (New York: Bantam Books, 1968). NAACP, Beyond the Rodney King Story (Boston: Northeastern University Press, 1995). Human Rights Watch, Shielded From Justice: Police Brutality and Accountability in the United States (New York: Human Rights Watch, 1998). 40 William A. Westley, Violence and the Police (Cambridge: MIT Press, 1970). 41 The argument is implicit in the long-standing recommendations that police departments diversify their work forces and actively recruit more racial and ethnic minority officers. See, for example, National Advisory Commission on Civil Disorders, Report, p. 315, 42 Catherine H. Milton, Women in Policing (Washington: The Police Foundation, 1970). 43 This view is central to the movement for police professionalism and the reform proposals that dominated the professionalization from about 1900 to the present. Samuel Walker, A Critical History of Police Reform (Lexington: Lexington Books, 1977). 44 On using peer officers to evaluate officers' performance, see Special Counsel, 16th Semiannual Report, p. 74. 45 Samuel Walker, Internal Benchmarking for Traffic Stop Data: An Early Intervention Approach. Discussion Paper. Omaha, NE, 2003. Available at www.policeaccountability.org. 46 Early Intervention State of the Art Conference, Report. 47 Dennis J. Kenney, Early Warning Systems: Responding to the Problem Police Officer (Washington: Government Printing Office, 2001), described a three-component approach. The four-component approach used here reflects the subsequent research. Available on the web at www.ncjrs.gov. NCJ 188565. 154 48 Early Intervention System State of the Art Conference, Report (Omaha, 2003). Available at www.policeaccountability.org. 49 A good description of these two systems is available in Special Counsel Merrick J. Bobb, 16th Semiannual Report (Los Angeles: Special Counsel, 2003), pp. 68–72. The report is available at www.parc.info. 50 Robert Davis, Christopher Ortiz, Nicole J. Henderson, Joel Miller and Michelle K. Massie, Turning Necessity into Virtue: Pittsburgh's Experience with a Federal Consent Decree (New York: Vera Institute of Justice, 2002), pp. 44–45. Available at www.vera.org. 51 Samuel Walker, Police Accountability: The Role of Citizen Oversight (Belmont: Wadsworth, 2001), pp. 121–135. 52 United States v. Los Angeles, Consent Decree (2001), Section II. Available at www.usdoj.gov/crt/split. 53 An important object lesson on this point is in Special Counsel Merrick J. Bobb, 16th Semiannual Report, which found that because of staff cutbacks data were not being entered into the systems in a timely fashion. The report is available at www.parc.info. 54 U.S. Department of Justice and the City of Cincinnati, Memorandum of Understanding, 2001. Available at www.usdoj.gov/crt/split. 55 Special Counsel Merrick J. Bobb, 16th Semiannual Report, pp. 71–72. 56 Early Intervention State of the Art Conference, Report. 57 Similar examples are offered in Samuel Walker, Internal Benchmarking for Traffic Stop Data: An Early Intervention System Approach. Discussion Paper (Omaha, 2003). Available at www.policeaccountability.org. 58 This approach is described in Bobb, 16th Semiannual Report, pp. 74–75. 59 This approach is described in ibid, pp. 68–69. 60 Walker, Alpert and Kenney, Early Warning Systems: Responding to the Problem Police Officer. 61 Special Counsel Merrick J. Bobb, 15th Semiannual Report (Los Angeles: Special Counsel, 2002), p. 39. Available at www.parc.info. 62 Ibid, p. 42. 63 Davis, et. al., Turning Necessity into Virtue, p. 46. 64 Commission on Accreditation for Law Enforcement Agencies, Standard 35.1.15. Personnel Early Warning System. 65 Special Counsel Merrick J. Bobb, 15th Semiannual Report, pp. 37–75. 66 New Orleans Police Department, Professional Performance Enhancement Program (PPEP) (New Orleans: NOPD, 1998). 67 Omaha Police Union, "Bad Boy/Girl Class Notes Shared." The Shield. (April 1992). 68 Early Intervention State of the Art Conference, Report. 69 San Jose Independent Police Auditor, 2001 Year End Report (San Jose: Independent Police Auditor, 2002), pp. 48–49. Available at www.ci.san-jose.us. The San Jose program is also described in Special Counsel Merrick J. Bobb, 16th Semiannual Report, p. 79. 70 Davis, et. al., Turning Necessity into Virtue. 71 New Orleans Police Department, Professional Performance Enhancement Program (PPEP). 72 Early Intervention State of the Art Conference, Report. 73 Comments, police chiefs, Justice Department Conference, Strengthening Police Community Relationships, June 910, 1999. 74 Herman Goldstein, Policing a Free Society (Cambridge: Ballinger, 1977), p. 171. 75 Hans Toch, J. Douglas Grant, and Raymond T. Galvin, Agents of Change: A Study in Police Reform (New York: John Wiley, 1975). 76 U.S. Commission on Civil Rights, Who is Guarding the Guardians? A Report on Police Practices (Washington, DC: U.S. Civil Rights Commission, 1981), pp. 159, 166. 77 Christopher Commission, Report of the Independent Commission on the Los Angeles Police Department (Los Angeles, 1991). Available at www.parc.info. 78 Ibid., pp. 40–48. 79 James G. Kolts and Staff, The Los Angeles County Sheriff's Department (Los Angeles: Los Angels County, 1992), pp. 157–167, 353. Available at www.parc.info. 80 Special Counsel Merrick J. Bobb, 11th Semiannual Report (Los Angeles: Special Counsel, 1999), pp. 46–47. Available at www.parc.info. 81 "Kansas City Police Go After Their 'Bad Boys'," The New York Times (September 10, 1991). 82 "Wave of Abuse Laid to a Few Officers," Boston Globe (October 4, 1992). 83 Sari Horowitz, "Officers Go Too Far," The Washington Post (November 19, 1998), p. 1. 84Samuel Walker, Geoffrey P. Alpert, and Dennis J. Kenney, Early Warning Systems: Responding to the Problem Police Officer (Washington, DC: National Institute of Justice, 2001). NCJ 188565. Available at www.ncjrs.gov. 85Catherine H. Milton, et. al., Police Use of Deadly Force (Washington, DC: The Police Foundation, 1977). Walker, Alpert, and Kenney, Early Warning Systems: Responding to the Problem Police Officer. 86 U.S. Commission on Civil Rights, Who is Guarding the Guardians?, pp. 81–86. 87 New York City, Commission to Investigate Allegations of Police Corruption and the Anti-Corruption Procedures of the Police Department [Mollen Commission], Commission Report (New York, 1994). 88 Tony Pate, Jack W. McCullough, Robert A. Bowers, Amy Ferrara, Kansas City Peer Review Panel: An Evaluation (Washington, DC: The Police Foundation, 1976). 89 George L. Kelling, et al., The Kansas City Preventive Patrol Experiment (Washington: The Police Foundation, 1974). 90 Pate, et al., Kansas City Peer Review Panel, pp. 9–10. 91 Ibid., p. 55. 92 "Kansas City Police Go After Their 'Bad Boys'," New York Times (September 10, 1991). 93 Toch, Grant and Galvin, Agents of Change. 94 U. S. Commission on Civil Rights, Who is Guarding the Guardians?. 95 John S. Ross, Memorandum to Chief Kenneth I. Harms, "Citizen Complaints Against Police Officers," (May 29, 1979), p. 1. 97 Ibid., pp. 2–3. 98 Ibid., p. 7. 99 Ibid., p. 10. 100 Ibid., p. 12. 101 Miami Police Department, Departmental Order, #2, Chapter 8. 102 Ibid. 103 "Miami Police Face Series of Inquiries After Shootings," The New York Times (July 18, 2001). 104 Bruce Porter, The Miami Riot of 1980 (Lexington: Lexington Books, 1984). U.S. Commission on Civil Rights, Confronting Racial Isolation in Miami (Washington: Government Printing Office, 1982). 105 Metro [Miami]-Dade Police Department, Professional Compliance Bureau, Early Identification System (internal memorandum). 106 Walker, Alpert, and Kenney, Early Warning Systems: Responding to the Problem Police Officer. 107 Ibid., 108 Samuel Walker, Police Accountability: The Role of Citizen Oversight (Belmont, CA: Wadsworth, 2001). 109 U.S. Department of Justice. Conference on Police Integrity. Report, 1997. 110 U.S. Department of Justice, COPS Office, Law Enforcement Working Session on Police Practices, November 29-30, 2000. 111 U.S. Justice Department, Principles for Promoting Police Integrity (Washington, DC: U.S. Justice Department, 2001). NCJ 186189. Available at www.ncjrs.gov. 112 United States v. Los Angeles, Consent Decree (2001). Available at www.usdoj.gov/crt/split/. 113 United States v. Cincinnati, Memorandum of Understanding (2002). Available at www.usdoj.gov/crt/split. 114 Commission on Accreditation for Law Enforcement Agencies (CALEA), Standard 35.1.15 (2001). Available at www.calea.org. 115 The first published literature are the reports and articles co-authored by the author of this report: Walker, Alpert, and Kenney, Early Warning Systems. 116 Ibid. 117 Special Counsel Merrick J. Bobb, 11th Semiannual Report, p. 55. 118 A full account is available in Special Counsel Merrick J. Bobb, 11th Semiannual Report, pp. 45–56. James G. Kolts, The Los Angeles County Sheriff's Department (Los Angeles: Los Angeles County, 1992). 119 Special Counsel Merrick J. Bobb, 11th Semiannual Report. 120 Special Counsel Merrick J. Bobb, 16th Semiannual Report, pp. 43–59, 67–81. 121 The chronology of events is very well documented in Los Angeles Police Commission, Report of the Rampart Independent Review Panel (Los Angeles: Los Angeles Police Commission, 2000), pp. 136–162 and Appendix B, pp. 217–219. 122 Christopher Commission, Report of the Independent Commission on the Los Angeles Police Department (Los Angeles: City of Los Angeles, 1991). 123 Merrick Bobb, Five Years Later: A Report to the Los Angeles Police Commission (Los Angeles: Los Angeles Police Commission, 1996). 124 Los Angeles Police Department, Board of Inquiry into the Rampart Area Corruption Incident (Los Angeles: LAPD, 2000). 125 Los Angeles Police Commission, Report of the Rampart Independent Review Panel. 126 Erwin Chemerinsky, An Independent Analysis of the Los Angeles Police Department's Board of Inquiry Report on the Rampart Scandal (Los Angeles, 2000). 127 United States v. Los Angeles, Consent Decree (2000). The consent decree is available on the DOJ website: www.usdoj.gov/crt/split. 128 United States v. City of Pittsburgh, Consent Decree (1997). Available at: www.usdoj.gov/crt/split/. 129 Discussed in Samuel Walker, "The New Paradigm of Police Accountability: The U.S. Justice Department 'Pattern or Practice' Suits in Context," St. Louis University Public Law Review, XXII (No. 1, 2003): 3–52. 130 Robert C. Davis, Christopher Ortiz, Nicole J. Henderson, Joel Miller and Michelle K. Massie, Turning Necessity into Virtue: Pittsburgh's Experience with a Federal Consent Decree (New York: Vera Institute of Justice, 2002), p. 44. Available at www.vera.org. 131 City of Pittsburgh, Monitor's Eighteenth Quarterly Report. Quarter Ending February 16, 2002. 132 Davis, et al., Turning Necessity into Virtue: Pittsburgh's Experience with a Federal Consent Decree. 133 Davis, et al., Turning Necessity into Virtue, p. 15. 134 Walker, Alpert, and Kenney, Early Warning Systems. 135 The evidence is summarized in Samuel Walker and Charles M. Katz, The Police in America: An Introduction, 4th ed. (New York: McGraw-Hill, 2003), Chs. 13, 14. 136 National Center for Women and Policing, Men, Women and Police Excessive Force: A Tale of Two Genders (Los Angeles: NCWP, 2002). Available at www.feminist.org. 137 Davis, et. al., Turning Necessity into Virtue, p. 49-50. 138 Special Counsel Merrick J. Bobb, 15th Semiannual Report (Los Angeles: Special Counsel, 2002), p. 68. Available at www.parc.info. 139 Ibid., p. 64. 140 Special Counsel Merrick J. Bobb, 16th Semiannual Report. 141 Early Intervention State of the Art Conference (February 2003), available at www.policeaccountability.org. 142 Walker, Alpert, and Kenney, 2001 143 The first effort in this regard was the Early Intervention System State of the Art Conference, held in Phoenix, Arizona, in February, 2003. See the conference Report, available at www.policeaccountability.org. Image of United States Department of Justice seal For More Information: U.S. Department of Justice Office of Community Oriented Policing Services 1100 Vermont AVenue, N.W. Washington, D.C. 20530 To obtain details on COPS programs, call the U.S. Department of Justice Response Center at 800.421.6770 Visit COPS Online at the address listed below. www.cops.usdoj.gov Created: August 25, 2003 CM#: e07032003 ISBN: 1-932582-26-6