Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
learning_to_build_police-community_trust_1
RESEARCH REPORT Learning to Build Police-Community Trust Implementation Assessment Findings from the Evaluation of the National Initiative for Building Community Trust and Justice Jesse Jannetta Sino Esthappan Jocelyn Fontaine Mathew Lynch Nancy La Vigne with Carla Vásquez-Noriega, Erica Kouka, Anamika Dwivedi, Ellen Paddock, and Dean Obermark August 2019 JUSTICE POLICY CENTE R ABOUT THE URBAN INST ITUTE The nonprofit Urban Institute is a leading research organization dedicated to developing evidence-based insights that improve people’s lives and strengthen communities. For 50 years, Urban has been the trusted source for rigorous analysis of complex social and economic issues; strategic advice to policymakers, philanthropists, and practitioners; and new, promising ideas that expand opportunities for all. Our work inspires effective decisions that advance fairness and enhance the well-being of people and places. Copyright © August 2019. Urban Institute. Permission is granted for reproduction of this file, with attribution to the Urban Institute. Cover image via Shutterstock. Contents Acknowledgments v Executive Summary vi Training viii Reconciliation xi Policy Change xii Chapter 1: The National Initiative 1 National Initiative Training and Technical Assistance Structure 3 The Evaluation of the National Initiative 4 The National Initiative Sites 7 Birmingham, Alabama 9 Fort Worth, Texas 10 Gary, Indiana 11 Minneapolis, Minnesota 12 Pittsburgh, Pennsylvania 13 Stockton, California 15 Baseline Levels of Community Trust in the Police 16 Chapter 2: National Initiative Trainings 18 The National Initiative Training Continuum 18 Implementation Experiences and Fidelity 20 Preparing for the Trainings 20 Training Delivery 21 Officer Responses to the Trainings 22 Community-Facing Training 34 Sustaining and Institutionalizing the Training Concepts 35 Chapter 3: Reconciliation 37 The National Initiative Reconciliation Process 37 Reconciliation Process Implementation 40 Fact-Finding 40 Acknowledgement of Harm 42 Sustained Listening: Reconciliation Listening Sessions 44 Narrative Collection and Dissemination 51 Chapter 4: Changing Police Department Policy 52 National Initiative Efforts That Inspired Policy Changes 52 IV CONTENTS Procedural Justice Model Policies 52 Policy Review and Analysis of Administrative Data 54 Training Feedback and Internal Procedural Justice 56 Reconciliation Listening Sessions 56 Group Violence Intervention 57 Policy Changes 58 Specific Policies 58 Formal Policy Review Processes 60 Institutionalizing the National Initiative Work within Police Departments 63 Chapter 5: Implementation Findings 65 Findings on Training Implementation 69 Findings on Reconciliation, Police Department Culture Change, and Sustainability 73 Chapter 6: Conclusion 79 Appendix A. National Initiative Site Timelines 82 Notes 88 References 93 About the Authors 96 Statement of Independence 97 ACKNOWLEDGMENTS V Acknowledgments This brief was supported by cooperative agreement number 2014-MU-MU-K051 awarded by the US Department of Justice, Office of Justice Programs, and Office of Juvenile Justice and Delinquency Prevention. We are grateful to them and to all our funders, who make it possible for Urban to advance its mission. The opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect those of the Department of Justice. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute’s funding principles is available at urban.org/fundingprinciples. The multiyear evaluation of the National Initiative benefitted tremendously from the talents and contributions of many Urban Institute colleagues past and present, and we want to particularly thank Isela Bañuelos, Joshua Eisenstat, Laura Golian, Emma Kurs, Dan Lawrence, Hanna Love, and Cameron Okeke. We also appreciate the assistance we received from the police department, residents, and community organizations in each National Initiative city, which was invaluable to collecting the information necessary for this report. We also wish to thank the National Initiative partners: the National Network for Safe Communities at John Jay College of Criminal Justice, the Justice Collaboratory at Yale Law School, and the Center for Policing Equity. VI EXECUTIVE SUMMARY Executive Summary Many communities throughout the United States that face high levels of crime and concentrated disadvantage—particularly communities of color—also struggle with high levels of mistrust in the police and strained police-community relations. Recognizing that a lack of legitimacy and community trust in policing was a serious and persistent problem with deep historical roots, and that addressing that problem required a well- resourced, multidimensional approach combining proven practices with new tools and approaches, the US Department of Justice launched the National Initiative for Building Community Trust and Justice. Led by John Jay College of Criminal Justice’s National Network for Safe Communities (NNSC), and in partnership with the Center for Policing Equity (CPE), Yale Law School (YLS), and the Urban Institute, the National Initiative for Building Community Trust and Justice (National Initiative) brought together practitioners and researchers to deliver a suite of interventions focused on law enforcement and community members in six cities: Birmingham, Alabama; Fort Worth, Texas; Gary, Indiana; Minneapolis, Minnesota; Pittsburgh, Pennsylvania; and Stockton, California. Core National Initiative interventions included (1) training and technical assistance for police officers on engaging with residents in a procedurally just manner, (2) trainings that encouraged officers to understand and mitigate implicit biases, (3) developing model police department policies and identifying key changes to extant policies, and (4) reconciliation discussions, during which police officers and community members had authentic conversations to acknowledge historic tensions, harms, and misconceptions and to repair relationships. The Urban Institute evaluated the National Initiative’s implementation and impact to inform potential replications and/or modifications of the initiative’s components, and to guide future research on police efforts to build community trust. The evaluation focuses on National Initiative activities occurring from January 2015 through December 2018. Researchers collected the following qualitative and quantitative data to support the evaluation: monthly teleconferences among members of the National Initiative implementation team that included partners from CPE, NNSC, and YLS publicly available information and media coverage of the National Initiative and issues pertaining to police-community relations in the pilot sites EXECUTIVE SUMMARY VII fieldwork that included observations of National Initiative activities and interactions between National Initiative partners and site stakeholders routine teleconferences with site coordinators, police chiefs, and other stakeholders documents provided by the sites and National Initiative partners semistructured interviews with police and community stakeholders in each site learning assessment surveys of officers receiving National Initiative trainings in each site surveys of residents in areas with high levels of concentrated crime and poverty/disadvantage in each site The implementation evaluation focused specifically on the successes and challenges of the collaboration among the National Initiative partners, participating police departments, and communities. Key findings included the following: The National Initiative partners had to overcome skepticism that outside experts were well- informed about local contexts and could provide new insights to strengthen police-community relationships. Agreeing on the presence and scope of the problem was a challenge in some departments. Successfully implementing the National Initiative required committed and skilled local site coordinators and distributing the work to other core teams (such as the procedural justice training team). Establishing consistent communication among the National Initiative team, police departments, and communities was difficult, and turnover within the National Initiative team in the early stages of the project exacerbated this challenge. Pairing the six sites with peer communities was a key facilitator of success. Sites benefitted from exchanges with other National Initiative cities and from consulting with their peer sites as well as other cities, such as Chicago, whose police department’s academy trainers helped implement training. Peer exchanges also helped with site partner morale, consistency, and innovation. Procedural justice (PJ) was the primary concept for orienting National Initiative activities within the police departments: all three trainings were named with the PJ shorthand, and organizational structures created to advance the work used names such as “the PJ unit.” VIII EXECUTIVE SUMMARY Community involvement in the National Initiative developed more slowly than police involvement, in part because community-focused components like the reconciliation process and the community-facing trainings were implemented at later stages, but also because communication with the community was not as strong as police and community stakeholders desired. Because training was one of the first National Initiative components fielded in the sites and was so resource intensive, the National Initiative focused on police departments in its early stages. Only later, as the community trainings and reconciliation process began, was there more robust community engagement. Training Training police officers in the concepts of procedural justice and implicit bias was a foundational component of the National Initiative, and was intended to promote trust-building interactions between officers and the public. Training involved nearly 100,000 person-hours and was the most resource- intensive National Initiative intervention in the participating police departments. Officers received substantial exposure to the National Initiative concepts, and training surveys and stakeholder interviews indicated widespread (though not universal) receptivity to the concepts. Practitioners also developed innovations during training implementation, including community-facing versions of the trainings to teach residents the core procedural justice and implicit bias concepts. The procedural justice and implicit bias training consisted of three full-day (eight-hour) segments on the following concepts: Procedural Justice and Police Legitimacy (conceptual procedural justice, or CPJ1) focuses on how procedural justice enhances community cooperation with the police and supports effective policing. It also explores how officers can incorporate procedural justice principles into their daily activities. A Tactical Mindset: Procedural Justice and Police Legitimacy (tactical procedural justice, or TPJ) is a more tactical approach to procedural justice, using simulations and scenario-based exercises and presenting techniques for applying procedural justice on the job.2 Tactical Perception: The Science of Justice (implicit bias, or IB) familiarizes officers with several key psychological concepts related to bias and discrimination, teaches them to identify how those concepts manifest in the world, and allows them to practice mitigating them. EXECUTIVE SUMMARY IX Officers who participated in these trainings were statistically significantly more likely to agree with procedural justice principles in the post-TPJ learning assessment than in the pretraining assessment see table 1). In general, these increases in levels of agreement were evident by the time CPJ concluded and were maintained through the end of TPJ. We observed increases in levels of agreement with procedural justice in all six National Initiative sites, with modest variation in the initial levels of agreement and degree of change. TABLE 1 Mean Differences on Procedural Justice Items Pre-CPJ Post-CPJ Post-TPJ Officers should at all times treat people they encounter with dignity and respect 3.59 3.69** 3.73** It is important to give everyone a good reason why we are stopping them 3.39 3.54** 3.55 Listening and talking to people is a good way to take charge of situations 3.41 3.60** 3.63* Police have enough trust in the public for them to work together effectively 2.46 2.84** 2.91** If people ask why we are treating them as we are, we should explain our actions as soon as we can 3.20 3.47** 3.56** When dealing with citizens' concerns, officers need to explain what will happen next when they are done at the scene 3.27 3.49** 3.50 People should be treated with respect, regardless of their attitude 2.97 3.24** 3.28 Officers need to show an honest interest in what people have to say, even if it is not going to do anything 3.23 3.46** 3.49 Officers should treat citizens as if they can be trusted to do the right thing 2.93 3.20** 3.25** It is important that we understand, follow, and protect the rights of the people with whom we come into contact 3.62 3.70** 3.76** It is very important that officers appear neutral in the application of legal rules 3.58 3.73** 3.74 Total procedural justice scale 3.24 3.46** 3.49** Notes: Valid n = 11,056. Cronbach’s alpha = 0.9. Response options ranged from 1 (strongly disagree) to 4 (strongly agree). Statistically significant differences are indicated for post-CPJ relative to pre-CPJ, and post-TPJ relative to post-CPJ, and were assessed using t-tests; *p < 0.05, **p < 0.01. Because Stockton police officers had already conducted a CPJ training before participating in the National Initiative, preassessment surveys were not fielded in that site. Additionally, postassessment surveys for the TPJ training were not fielded in Gary or Pittsburgh. The implicit bias learning assessment surveys found that officers’ views were better aligned with the concepts after the training than before. This was particularly true for items related to how stereotypes can influence individual behavior and the need to mitigate that influence through reflection and awareness (figure 1). X EXECUTIVE SUMMARY FIGURE 1 Officer Agreement with Implicit Bias Concepts Note: *p < 0.05, **p < 0.01. Key findings from the assessment of training implementation included the following: Ensuring that procedural justice trainers were “credible messengers” was a key to success. This was necessary for overcoming police officers’ resistance to charged material such as the racist history of policing, understanding the perspectives of community members who distrusted the police, and implicit biases. Delivering 24 hours of training to every sworn officer in a police department placed a heavy load on the trainers and the departments. This strain was greater for trainers who retained other job responsibilities in addition to delivering the trainings. Extending the trainings to the community was an important innovation, but using uncompensated volunteer community trainers was a barrier to doing so. Issues with internal procedural justice (i.e., the application of procedural justice within police departments) could impede officer uptake of the PJ concepts in their interactions with community members. 3.22 3.78 2.49 3.24 2.72 2.73 3.34 3.78 3.19 4.04** 2.28** 3.30* 2.79* 3.02** 3.58** 4.09** My biases do not affect my behavior Even well-meaning people have biases If implicit biases are unconscious, then there's nothing I can do about them I am worried that someone will prejudge me because I am a police officer My concern about being prejudged as a police officer could impact my behavior Stereotypes about particular groups could influence my interactions with them, without my awareness At times, situations can influence our decisions and behaviors, more than personal character Being reflective and aware of how I respond to people and situations can increase my effectiveness as a police officer Pretraining Posttraining EXECUTIVE SUMMARY XI Reconciliation Although reconciliation was one of the National Initiative’s three conceptual “pillars,” a specific reconciliation process did not exist when the initiative began. Of the three pillars, reconciliation was the most nascent and previously untested component. National Network for Safe Communities developed a framework for a police-community reconciliation process (figure 2). Though it was implemented in all of the National Initiative sites, they made various degrees of progress implementing the framework’s components. FIGURE 2 National Initiative Reconciliation Framework The reconciliation framework consists of the following five key components: Fact-finding is conducted regarding police departments’ past harms (such as enforcing Jim Crow laws) and present harms maintained through policies and practices with detrimental effects on safety and justice. Fact-finding informs both the historical components of the procedural justice trainings for police officers and departments’ public acknowledgements of harm. Fact-finding on present problems provides a basis for identifying changes and reparative actions that the police department can adopt to better align with community priorities and build community confidence in the overall process. XII EXECUTIVE SUMMARY Acknowledgements of harm by police leadership involve recognizing the police’s historical and present harms, as well as ongoing problems that fuel mistrust between the police and community. Such harms may include officer-level issues (e.g., disrespectful treatment of civilians) and department-level issues (e.g., overuse of stop and frisk tactics). Sustained listening is facilitated via listening sessions, where police provide an intimate, nonadversarial forum for community members to share their experiences with and insights about law enforcement. These sessions occur in small settings (typically with 8 to 20 participants) and are a key mechanism for identifying narratives and informing specific changes to policy and practice that are then reported back to community members in subsequent listening sessions. Narratives are collected and shared to capture communities’ common perceptions of police and the police’s common perceptions of communities. Some of these narratives may illuminate common ground for joint action, such as a shared commitment to reducing violence. However, others may highlight impediments to police-community reconciliation (for example, some law enforcement officers believed that communities were tolerant of crime and violence, and some community members believed that the police deliberately engaged in harmful actions). These narratives are shared with the public to build mutual understanding and empathy among a broader group of people. Explicit commitments to changing policy and practice are made in areas identified through the listening sessions. This means making concrete, reparative, and substantive modifications to department policy and practice. Resulting changes are communicated to reconciliation process participants and the public. Making and communicating these changes are both necessary for creating confidence that the reconciliation process can effect meaningful change. Policy Change The National Initiative was based on the logic that sustaining public trust in police requires new thinking and practice. Making and sustaining changes to organizational culture required changing individual police officers’ thinking and practices and institutionalizing new ways of policing in departmental policies and structures. Police stakeholders stressed in interviews that developing policy was important for supporting the consistency and long-term sustainability of practices intended to build trust. Table 2 provides a summary of policy changes resulting from or influenced by National Initiative activities. EXECUTIVE SUMMARY XIII TABLE 2 Policy Changes during the National Initiative Implementation Period, by Department Policy changes City Birmingham designated a sergeant as a liaison to the LGBTQIA+ community (September 2016) modified policy language to explicitly reinforce commitment to unbiased policing (2017) protection from abuse orders provided to all precincts by court rather than stored only in the precinct where the order was awarded, or rather than survivors being responsible for providing the order (2017) created new command-level position overseeing all community engagement (October 2018) Fort Worth revised general order on bias-free policing (February 2016) created standard operating procedures for Procedural Justice Unit (November 2016) began reporting use-of-force, arrest, stop, and discipline policies and statistics online (2017) revised general order on sexual assaults (July 2017) issued new order on racial profiling that reaffirmed the department’s commitment to unbiased policing (January 2018) revised use-of-force/force options and use-of-force reporting general orders (March 2018) revised departmental mission statement to add commitment to respect the sanctity of human life and preserve the rights and dignity of each individual in the community (March 2018) created new Police and Community Relationships general order, including role of Procedural Justice Unit (July 2018) added a duty to protect the safety and physical health of arrested and detained persons to the department’s Arrest Procedures General Order (August 2018) Gary None Minneapolis added transgender/gender nonconforming policy (June 2016) amended use-of-force policy to prioritize sanctity of life for both officers and civilians (July 2016) added policy requiring officers to intervene in incidents in which other officers use excessive force (July 2016) began tracking race and gender on traffic stops and other stops (September 2016) changed body-worn camera policy to require officers to turn on cameras as soon as they begin responding to 911 calls (July 2017) began reporting officer use-of-force, complaint, stop, crime, and arrest statistics online (2017) failure by an officer to comply with a lawful investigation of misconduct shall be deemed an act of misconduct (September 2018) Pittsburgh created ethics document (2016) added procedural justice concepts to evaluations of field training officers and recruits (2016) added order on transgender and gender nonconforming employees (August 2016) began posting policies online (January 2018) Stockton added procedural justice language to general order on how canines are deployed (May 2016) modified field training officer evaluations to add demonstration of PJ practices, and required supervisors to evaluate officer understanding of PJ in considering transfer/special assignment requests and promotional examinations (2016) mandated that officers receive annual mental health training mandated that officers make every attempt to mediate and defuse situations with people experiencing a mental health crisis (July 2016) added policy stating that “sworn personnel of the Stockton Police Department shall not stop, question, detain, arrest or place ‘an immigration hold’ on any person solely on the ground that he or she may be a deportable alien” (January 2017) added procedural justice language to rules and regulations regarding conduct toward the public and fellow police members (May 2017) tenets of procedural justice added to equestrian unit order (June 2017) created policy on release of body camera footage (July 2017) Unmanned Aircraft System policy created with community input (November 2017) XIV EXECUTIVE SUMMARY The implementation assessment’s key findings regarding reconciliation, policy change, and sustaining National Initiative activities in the sites included the following: Successful reconciliation listening sessions required police to be open to hearing community perspectives and not to react defensively to critical or emotionally charged statements. Police leadership, most notably from the chief, was necessary to support and model the National Initiative. Leadership support is critical for teaching procedural justice to officers, speaking for the department in reconciliation activities, making key changes to policy, and providing the organizational support necessary to advance the National Initiative’s work. Changing police chiefs could interrupt progress. Turnover in chiefs is common, and four of the six National Initiative sites saw a new chief during the implementation period. Such changes particularly disrupted the reconciliation and policy change work, delaying progress even when the new chief supported the work. Deliberately selecting the training teams also created internal champions of National Initiative concepts. In some departments, trainers became part of designated procedural justice units, and some members of the training teams became involved in other National Initiative activities such as community trainings, reviews of department policy , and reconciliation listening sessions. Chapter 1: The National Initiative Many communities throughout the United States that face high levels of crime and concentrated disadvantage—particularly communities of color—also struggle with high levels of mistrust in the police and strained police-community relations. In recent years, this lack of trust has been the subject of protest and public conversation, prompted by a series of police encounters that resulted in the deaths of primarily African American men, women, and boys. Public discourse around changing policing has elevated awareness of the critical need to enhance accountability for police use of force; reassess broken windows” tactics that emphasize heavy enforcement of low-level offenses in neighborhoods experiencing high rates of crime and poverty; and reduce racial and ethnic disparities in police contact, arrests, and incidents involving the use of force. Although the US justice system is charged with increasing safety and reducing victimization, many of the activities its law enforcement arm undertakes in the nation’s most economically and socially distressed communities burden residents. Ranging from fees and fines from enhanced traffic and misdemeanor enforcement to high rates of incarceration, these burdens are particularly borne by people of color, who are disproportionately represented across the justice continuum (Hartney and Vuong 2009). In the worst cases, exemplified in Ferguson (DOJ 2015) and Baltimore (DOJ 2016), abusive policing practices can trigger civic unrest. This situation could be characterized as a crisis in the legitimacy of police and policing in the communities that most need protection. Legitimacy in policing is the belief that “the police ought to be allowed to exercise their authority to maintain social order, manage conflicts, and solve problems in their communities” (Tyler 2004, 9). Public belief in the legitimacy of the police (and the justice system generally) has benefits including greater compliance with the law and willingness to help authorities reduce crime and maintain order (Tyler, Goff, and MacCoun 2015). Cooperation between law enforcement agencies and the communities they serve can advance safety and justice by, for example, enabling members of the public to provide law enforcement with information about crimes. Greater police legitimacy may also support officer and resident safety by making their encounters less tense or less likely to escalate. The US Department of Justice (DOJ) launched the National Initiative for Building Community Trust and Justice (National Initiative) after recognizing that a lack of legitimacy and community trust in policing was a huge, persistent problem with deep historical roots, and that addressing that problem required a well-resourced, multidimensional approach combining proven practices with new tools and 2 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS approaches. Established through a cooperative agreement from the DOJ in 2014, the National Initiative piloted interventions in six cities to restore relationships between police and communities suffering from high levels of crime and strained police-community relations. Led by John Jay College of Criminal Justice’s National Network for Safe Communities (NNSC), and in partnership with the Center for Policing Equity (CPE), Yale Law School (YLS), and the Urban Institute, the National Initiative brought together practitioners and researchers to deliver a suite of interventions focused on law enforcement and community members in six cities: Birmingham, Alabama; Fort Worth, Texas; Gary, Indiana; Minneapolis, Minnesota; Pittsburgh, Pennsylvania; and Stockton, California.3 To achieve its mission, the National Initiative was built upon three core areas or pillars that research evidence and practical experience suggested could generate measurable changes in police- community interactions and policing practices. The first pillar, procedural justice, refers to a set of practices that authorities can employ to increase their legitimacy with members of the public. The four essential tenets of procedural justice are (1) affording all people dignity and respect (respect), (2) assuring that all perspectives are heard during interactions (voice), (3) making decisions impartially neutrality), and (4) demonstrating trustworthy motivations for one’s actions (trustworthiness) (Tyler 2004). The procedural justice pillar comprised two categories: conceptual procedural justice (CPJ) and tactical procedural justice (TPJ). The second pillar, implicit bias (IB), focuses on how unconscious biases shape criminal justice agents’ actions and lead to racially disparate outcomes even when those actions are not overtly or consciously racist. In other words, it refers to the automatic associations that people make that influence their behavior and judgments. A large body of research focuses on racially disparate outcomes in law enforcement practices such as pedestrian and vehicle stops and the disproportionate use of force Eith and Durose 2011; Glaser 2014; Jones-Brown, Gill, and Trone 2010; Pierson et al. 2017; Seguino and Brooks 2018). Numerous studies suggest these outcomes may be attributable to implicit biases, stereotypes, or other psychological mechanisms that link people of color, and especially Black individuals, with crime and violence (Ross 2015). The third pillar, reconciliation, focuses on how frank conversation and open acknowledgement of historic tensions, harms, and misconceptions can repair relationships and lay a foundation for building trust between law enforcement agencies and the communities they serve. The National Initiative’s reconciliation process was designed to engage communities and police to acknowledge law enforcement’s past and present harms, allow communities to express their views authentically (and sometimes emotionally), and change policing practices to better reflect shared understandings of community safety. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 3 This report presents findings from Urban’s implementation assessment of the National Initiative. The remainder of this chapter provides an overview of the National Initiative’s technical assistance structure, presents Urban’s implementation evaluation approach, and describes the sites and the challenges they faced around police-community trust. Chapters 2, 3, and 4 are each devoted to a key component of the National Initiative intervention package: changing the nature of individual-level interactions between police officers and civilians, primarily through a series of officer trainings (chapter 2); creating positive police-community relationships through a defined reconciliation process (chapter 3); and changing police departments’ cultures and organizational structures, primarily through changes to policy and practice (chapter 4). These chapters discuss how the National Initiative designed these components, how they were implemented, and how implementation varied across sites. The report then presents stakeholder perspectives and reflections on implementation (chapter 5) and concludes with key takeaways (chapter 6). National Initiative Training and Technical Assistance Structure The National Initiative implementation package was oriented around the three pillars, and the partner organizations each oversaw one pillar (figure 1.1). The National Initiative faced the complex challenge of structuring partnerships among numerous partner organizations and stakeholders in each participating city—including law enforcement, government actors, and community stakeholders and residents—to carry out the core interventions. These interventions included (1) training and technical assistance for police officers on engaging with residents in a procedurally just manner, (2) trainings that encouraged officers to understand and address situations involving implicit biases, (3) developing model police department policies and identifying key changes to extant policies, and (4) reconciliation discussions, during which police officers and community members had authentic conversations to acknowledge historic tensions and harms and repair relationships. National Network for Safe Communities, which housed the National Initiative’s project director, bore responsibility for facilitating interactions between the National Initiative team and the sites. Sites were selected in March 2015, and the National Initiative team began partnerships with each site at kickoff visits in spring 2015 and developed those relationships through the end of that year. Representatives from the National Initiative partner organizations and the DOJ participated in those visits to build connections with the police, local government leadership, and key community organizations and constituencies, explain the National Initiative’s purpose and its three conceptual 4 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS pillars, and identify immediate next steps to begin implementation. Initial implementation plans were shared with each site and were the focus of a second site visit that NNSC conducted in late 2015. At each site, the DOJ funded a site coordinator,4 a key figure who served as the primary liaison with the National Initiative partners. The six site coordinators played a central role in driving implementation and developing partnerships and connections with potential community partners. Site coordinators worked either within or in close partnership with the police departments. Coordinators at four sites were (or had been) sworn officers: Birmingham’s was a retired captain, Fort Worth’s first coordinator was a lieutenant (an assistant chief replaced him when he stepped out of the role), Pittsburgh’s was a commander, and Stockton’s was a lieutenant. The Minneapolis site coordinator was a civilian employee of the police department, and Gary’s was the civilian coordinator of the city’s violence reduction model. FIGURE 1.1 National Initiative Structure The Evaluation of the National Initiative The Urban Institute evaluated the National Initiative’s implementation and impact to inform replications of and/or modifications to National Initiative components, and to guide future research on NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 5 law enforcement’s community trust-building efforts. Urban researchers collected primary and secondary data on activities at two levels: the National Initiative level (e.g., the development and implementation of training and technical assistance, site visits, and peer exchanges) and the site level. The following questions guided the implementation evaluation: Were National Initiative activities designed and implemented as planned? Were the National Initiative training and technical assistance activities effective in transmitting information? What interventions were designed and implemented? What interventions were designed and implemented specific to the target groups? The evaluation focused on National Initiative activities undertaken from January 2015 through December 2018. Researchers collected the following qualitative and quantitative data to support the evaluation: monthly teleconferences among the National Initiative implementation team that included partners from CPE, NNSC, and YLS publicly available information and media coverage of the National Initiative and issues pertaining to police-community relations in the pilot sites fieldwork that included observations of National Initiative activities and interactions between National Initiative partners and site stakeholders routine teleconferences with site coordinators, police chiefs, and other city stakeholders documents provided by the sites and National Initiative partners semistructured interviews with police and community stakeholders in each site learning assessment surveys of officers receiving National Initiative trainings in each site surveys of residents in areas with high levels of concentrated crime and poverty/disadvantage in each site Partner organizations provided technical assistance and engagement remotely, during in-person site visits, and during occasional peer learning visits to observe practices in non –National Initiative peer jurisdictions. In January 2016, NNSC began organizing routine teleconferences with the sites that CPE, YLS, and Urban participated in. There were also monthly check-in teleconferences with the site coordinator, calls with the chiefs of police, and “citywide” teleconferences that invited representatives 6 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS from the mayor’s office, local district attorney’s office, US Attorney’s Office, and the DOJ’s Community Relations Service (although only a subset of these agencies was present in many teleconferences).5 As the National Initiative was implemented, other key points of contact joined the site liaisons on the teleconferences, such as members of the procedural justice training teams and policy review teams and reconciliation session participants. In the latter half of 2018, the National Initiative hosted cross-site teleconferences for the procedural justice trainers and policy review teams to identify and troubleshoot common issues. Frequent site visits and peer exchanges among the sites complemented this remote support. After the initial site visits in 2015, these in-person visits and exchanges generally focused on planning the implementation of specific National Initiative activities, and typically involved the partner organizations responsible for each component (i.e., Yale Law School for CPJ and TPJ, Center for Policing Equity for IB, and NNSC for reconciliation). We discuss these activity-specific training and technical assistance efforts in greater detail in subsequent chapters. Urban researchers participated in many of these discussions and activities, and the evaluation draws on their observations about the substance and content of the meetings and participants’ interactions. The evaluation also relies on NNSC and CPE partners’ notes on their delivery of technical assistance to the sites. Teams of two Urban researchers periodically visited the sites and partner offices during the evaluation. During these visits, we convened with other partners to discuss the design and progress of National Initiative activities, observed the six sites’ procedural justice and implicit bias trainings, and observed partner interactions with site stakeholders and community members. We also conducted semistructured interviews with 123 individuals across all sites between 2016 and 2018. Interviewees included the partners at CPE, NNSC, and YLS, and key staff and stakeholders in the six sites, including police chiefs, procedural justice trainers, and site coordinators. The evaluation also included learning assessment surveys administered to officers who participated in the procedural justice and implicit bias trainings. To help the sites measure how well the trainings were received and make any necessary changes to the content or instructors, the researchers developed the learning assessment surveys in consultation with Yale Law School for CPJ and TPJ and Center for Policing Equity for IB. The PJ training teams in each site also had the opportunity to provide feedback before the learning assessments were administered. These surveys were self-administered on a paper-and-pencil instrument before and after the trainings and were designed to measure how well officers understood the concepts and their perceptions of the training and trainers. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 7 The procedural justice trainers administered their pretraining assessments before CPJ to gauge officers’ agreement with the importance of the concepts of voice, respect, neutrality, and transparency; at the end of the day’s training, they administered the posttraining survey with the same procedural justice items as well as questions regarding the quality of the training. Measures of procedural justice concepts were drawn from scales developed and validated by Skogan, Van Craen, and Hennessy (2015) in their evaluation of the Chicago Police Department’s procedural justice training. The posttraining survey was readministered after TPJ in Fort Worth, Gary, Minneapolis, and Pittsburgh. Birmingham completed CPJ and TPJ on consecutive days and administered the pretraining survey at the beginning of the first day and the posttraining survey at the end of the second day. Stockton began with TPJ and administered the pretraining survey before that class. The implicit bias pre- and posttraining surveys were administered at the beginning and end of the training day, respectively. In addition to the data collected for the implementation evaluation, researchers collected other data for the outcome and impact evaluation. These data included two waves of surveys of residents living in areas with high levels of concentrated crime and poverty/disadvantage in each site (wave 1 N = 1,278; wave 2 N = 1,202) and administrative records from each site’s police department. Additional description of the data collection process and findings from these data are described in two complementary reports (see Lawrence and colleagues [2019] for the impact evaluation findings and Fontaine and colleagues [2019] for the community survey findings). The National Initiative Sites Soon after then–Attorney General Eric Holder announced the National Initiative in September 2014, its partners began selecting sites by collectively identifying approximately 100 candidate police departments in consultation with the DOJ. The partners considered whether candidate departments were likely to have the commitment and capacity to effectively implement National Initiative components and serve as good sites for assessing the initiative’s impact. The intent was also to select sites that reflected the diversity of US cities as closely as possible so that other cities could learn from their efforts. Candidate departments were ones that the DOJ and National Initiative partners knew had worked on collaborative policing reform, procedural justice, implicit bias, reconciliation, or violence prevention, as well those that had expressed interest to the DOJ or National Initiative partners in being considered. Police departments that were subject to a federal consent decree or a federal pattern or practice investigation were excluded. 8 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS From November 2014 to January 2015, the National Initiative and DOJ partners winnowed the list of candidate sites iteratively by considering the capacity of each site’s police department and city leadership to undertake meaningful change, known issues around procedural justice, implicit bias, and racial tension, and promising ongoing efforts that could form the foundation for substantial progress. The partners also considered jurisdictional size, region, and visibility to cull a cohort of sites that would allow researchers to field test the National Initiative and its components in varied contexts and allow for comparisons among peer jurisdictions. By January 2015, the National Initiative and DOJ partners had identified 12 finalist sites. The National Initiative team communicated with these cities’ chiefs of police and leadership to gauge their commitment and interest. Urban collected descriptive information on crime rates and police department size from each city and had discussions with key personnel about their departments’ efforts that related to the National Initiative’s three pillars. This information complemented communication with the chiefs and city leadership to inform the final site selection. In March 2015, the National Initiative and DOJ partners selected Birmingham, Fort Worth, Gary, Minneapolis, Pittsburgh, and Stockton as the initial cohort of pilot sites. Timelines for each site are included in the appendix and illustrate key National Initiative implementation milestones through the end of 2018, key site-specific events, and events throughout the US that might have impacted police-community trust. TABLE 1.1 Descriptive Overview of National Initiative Sites Birmingham Fort Worth Gary Minneapolis Pittsburgh Stockton Population 212,237 812,238 80,314 382,578 305,704 302,389 African American (non-Hisp) 73.4 18.9 84.8 18.6 26.1 12.2 Hispanic/Latinx 3.6 34.1 5.1 10.5 2.3 40.3 Asian 1.0 3.7 0.2 5.6 4.4 21.5 2+ races 1.0 3.1 2.1 4.4 2.5 6.9 Native American 0.2 0.1 0.3 2.0 0.2 1.1 Median household income $31,061 $53,214 $28,020 $51,480 $40,715 $44,797 Sworn officers 838 1,681 178 847 885 441 Violent crime rate (per 100k) 1,588 487 913 1,012 798 1,331 Sources: “Comparing 2015 American Community Survey Data,” US Census Bureau, last updated June 17, 2018, https://www.census.gov/programs-surveys/acs/guidance/comparing-acs-data/2015.html; “Crime in the US 2015,” FBI, https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015 (crime rates and number of sworn officers for Birmingham); Crime in the US 2017,” FBI, https://ucr.fbi.gov/crime-in-the-u.s/2017/crime-in-the-u.s.-2017 (number of sworn officers in Fort Worth, Gary, Minneapolis, Pittsburgh, and Stockton). NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 9 Birmingham, Alabama Birmingham is a midsized city in north-central Alabama with a population of 212,461 that is predominantly African American (73.4 percent).6 Although crime rates declined steadily in Birmingham over the 10 years before it joined the National Initiative, the city experiences rates of violent crime well above the national average for cities with populations between 100,000 and 250,000.7 Concentrated poverty is also a serious issue in Birmingham: between 2010 and 2014, an estimated 58.4 percent of metro residents lived in neighborhoods where at least 20 percent of residents had incomes below the poverty level (Kneebone and Holmes 2015). With a legacy rooted in the civil rights movement, Birmingham is replete with community-based organizations seeking to uplift and repair communities damaged by its fraught history with policing. Birmingham has a structured network of 99 neighborhood associations that serves as the main apparatus through which residents communicate concerns to city government. The Birmingham Police Department had 838 sworn officers in 2015. Its chief at that time was A. C. Roper, who had served as chief since November 2007. Police-community relations in Birmingham are rooted in the BPD’s adversarial role during the civil rights movement and its long history of enforcing Jim Crow laws. The BPD has long struggled to engage community residents, particularly African Americans. Chief Roper had been unusually forthright in discussing the BPD’s historically fraught relations with African American people in Birmingham, including its officers’ enforcement of segregation and white supremacy. However, despite decades of efforts to alleviate tensions between Birmingham’s African American residents and its police department,8 interviews with Birmingham stakeholders (including police and community leaders) indicated lingering tensions between those communities. One issue that stakeholders raised in interviews and during the first site visit was that many BPD officers lived outside of Birmingham, which they perceived as distancing officers from residents. There were challenges within the police department as well: stakeholder interviews and discussions with BPD officers during the first site visit surfaced concerns about a perceived lack of transparency in internal BPD processes, particularly in hiring and promotion. The BPD joined the National Initiative having recently taken tangible measures to increase oversight and accountability, most notably by deploying body-worn cameras . In June 2015, Birmingham rolled out 319 body cameras to police precinct and task force officers. The BPD subsequently saw a 34 percent drop in use-of-force incidents and a 70 percent drop in citizen complaints regarding use of force.9 10 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Fort Worth, Texas Fort Worth, located in north-central Texas, has 812,238 residents and is the largest of the six National Initiative cities. With more than 1,500 police officers, the Fort Worth Police Department was also by far the largest department participating in the National Initiative. Moreover, Joel Fitzgerald became the department’s first African American chief in 2015, shortly before Fort Worth was selected as a National Initiative site. Like many US police departments, the FWPD’s officer demographics are not representative of its city’s residents. In 2014, 16 percent of officers were Hispanic/Latinx, compared with 34 percent of the city’s population, and 12 percent of officers were African American, compared with 18.9 percent of the population. The FWPD was navigating a challenging transition to new leadership tasked with addressing several organizational issues. Interview respondents indicated that a lack of communication and inconsistent discipline, which created a widespread sense of unfairness, were problems under the previous FWPD administration that needed to be addressed. Additionally, several African American FWPD officers had filed a federal suit against Chief Fitzgerald’s predecessor, Jeffrey Halstead, for racial discrimination and harassment.10 Thomas Windham, FWPD’s chief from 1985 to 2000, helped establish the department’s tradition of neighborhood policing practices, and many interviewees discussed the importance of his leadership in establishing positive relationships between FWPD and the community. Although Chief Halstead’s tenure was perceived to have undone some of this legacy, foundational practices from that time such as the Neighborhood Patrol Officer program were still in place when the National Initiative began. Interviews with stakeholders in Fort Worth indicated that Chief Fitzgerald was more engaged with the community than recent chiefs, and many suggested that he was the most engaged since Chief Windham. In 1995, Fort Worth residents voted to establish a Crime Control and Prevention District (CCPD) supported by a half-cent sales tax to fund crime prevention strategies. Since then, residents have consistently voted to renew CCPD legislation (as recently as 2014). Since 1995, Fort Worth's crime rate has declined 40 percent, while the city’s population has steadily increased (Fort Worth CCPD 2014). Interviewees cited the CCPD as an indication of residents’ support for and trust in the police department. Officers specifically noted that continued renewal of the CCPD speaks volumes about how much Fort Worth residents value their services. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 11 Gary, Indiana Gary, located in northwestern Indiana, has about 80,000 residents (the fewest of the six National Initiative pilot sites), 84.8 percent of whom are African American.11 Although Gary has seen a steady decline in crime over the past 25 years (consistent with national trends), it still experiences high rates of violent and property crime.12 Gary is well known for suffering extensive economic disinvestment and blight. In the four decades since US Steel disinvested in Gary, the city’s population has declined from 175,000 to 80,000. Approximately one-quarter of the buildings in Gary are abandoned.13 When National Initiative activities began, the Gary Police Department’s chief was Larry McKinley, a 17-year department veteran who assumed leadership in 2015 following a decade of instability that saw multiple chief changes. The GPD had 181 sworn officers in 2015, down from 222 in 2013 (Indiana State Police 2013). The Lake County Sheriff’s Department supplements GPD’s efforts as needed. Tensions between the police and community in Gary have long been an issue, exacerbated by several high-profile incidents. In 2009, for example, the DOJ Civil Rights Division investigated former chief Thomas Houston and two other members of the command staff for violating the civil rights of a Gary resident accused of burglarizing Houston’s home.14 Houston was sentenced to 41 months in prison and 2 years of supervised release for physically assaulting the handcuffed victim by kicking and striking different parts of his body. Moreover, GPD entered the National Initiative soon after Carl Blount Jr. had fatally shot Officer Jeffrey Westerfield in 2014.15 In 2013, the DOJ’s Diagnostic Center and the Indiana State Police collaborated to form the Gary Police Department Technical Assessment Team. In October 2013, the GPD Technical Assessment Team issued a report finding (1) a lack of direction and discipline within the GPD, (2) a critical deficiency in record keeping, and (3) inefficient allocation of police personnel to addressing community needs Indiana State Police 2013).16 The GPD also faced other organizational challenges that hurt morale, including low wages: GPD officers had not received a pay raise in more than 10 years when it joined the National Initiative.17 The GPD Technical Assessment Team’s report also expressed concerns about resource deprivation, such as the GPD’s limited uniform expenses, which sometimes made it difficult for officers to adhere to professional appearance standards (Indiana State Police 2013). During the National Initiative, community members routinely expressed concerns about the professionalism and appearance of GPD officers. 12 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Minneapolis, Minnesota Minneapolis, located in eastern Minnesota, has a population of 410,939 that immigration patterns have made much more racially and ethnically diverse in recent decades.18 Per capita income in Minneapolis is substantially higher than the US median. However, Minneapolis also has high levels of economic deprivation compared with other major US cities: 22.6 percent of city residents live below the poverty level (Berube and Holmes 2015). Similarly, although Minneapolis was ranked as the most literate city in the US in 2014,19 its 4-year graduation rates are among the country’s most racially disparate. In Minneapolis public schools, African American students graduate at rates 29.3 percent lower than white students, Hispanic or Latinx students 24.6 percent lower, and American Indian and Alaska Native students 45.5 percent lower.20 The Minneapolis Police Department employs 847 sworn officers. Janeé Harteau, who was chief when Minneapolis joined the National Initiative, had led the department since 2012 and was both the first woman and the first openly LGBTQIA+-identifying person to serve as the city’s police chief. There is a long history of tension between MPD and Minneapolis communities. The department’s history with Native American residents, for example, includes an incident involving the transportation of Charles Lone Eagle and John Boney in the trunk of a police car in 1993.21 Its histories with the Hmong and Somali communities have also been fraught, and include the 2006 shooting of 19-year-old Fong Lee and a 2015 video of an officer threatening to break 17-year-old Faysal Mohamad’s legs.22 That 94 percent of sworn officers live outside city limits may continue to contribute to such tensions.23 The MPD entered a federal mediation process in 2002 under the DOJ’s Community Relations Service after a police officer injured an 11-year-old with a stray bullet during a drug raid in north Minneapolis, leading to major protests and demands for federal intervention.24 The city, MPD, and the Unity Community Mediation Team ultimately signed a memorandum of agreement on December 4, 2003, that presented a series of action items related to use of force, police-community relations, mental health issues, department diversity, accountability, removal of children from their homes, and training.25 The agreement resulted in several policy changes; however, the city did not renew the memorandum of agreement when it expired in December 2008.26 In November 2013, Chief Harteau invited the Office of Justice Program’s Diagnostic Center to review oversight policies and practices and make recommendations for improvement.27 That review confirmed earlier reports that serious disciplinary action occurred infrequently, with the majority of sustained complaints resulting in interventions such as coaching (OJP Diagnostic Center 2015). In stakeholder interviews and initial National Initiative community meetings, community members repeatedly expressed disappointment NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 13 with past DOJ interventions and investigations in Minneapolis, viewing these efforts as ineffective and misguided. Arrests in Minneapolis declined 32 percent from 2007 through 2014,28 and proactive police stops, which include traffic law enforcement, suspicious person, and suspicious vehicle stops, dropped sharply in 2014.29 However, since 2000, two major investigations have found substantial racial disparity in MPD police activity. A 2003 statewide racial profiling report prepared for the Minnesota legislature analyzed traffic stops and found that in Minneapolis, African American residents were stopped 152 percent more often than would be expected if stops were consistent with the proportion of African American drivers in the overall driving population. Similarly, African American residents were searched 52 percent more often than white residents once stopped. Moreover, MPD officers stopped Latinx residents 63 percent more often than expected and conducted searches 15 percent more often (Council on Crime and Justice and Institute on Race & Poverty 2003). An ACLU investigation found that compared with whites, African American residents were 8.7 times more likely and Native American residents 8.6 times more likely to be arrested for low-level offenses.30 Stakeholder interviews and several protests before and during the National Initiative demonstrate Minneapolis residents’ high degree of civic engagement and desire to address issues around policing, trust, and public safety. Residents described a contentious relationship between the police department and community members and organizations, and a sense that MPD was looking to protect officers involved in use-of-force incidents (the November 2015 shooting of Jamar Clark was specifically cited). Minneapolis has several civic engagement groups that, though critical of MPD in the past, had potential as important focal points for community engagement efforts. Pittsburgh, Pennsylvania Pittsburgh is a midsized city in western Pennsylvania with a population of approximately 305,000.31 Pittsburgh residents are predominantly non-Hispanic white (64.8 percent) and African American (26.1 percent). The Pittsburgh Bureau of Police’s chief when the National Initiative began was Cameron McLay, who became chief in September 2014 after a year of interim leadership following the 2013 conviction of the previous chief on corruption charges. McLay was tasked with addressing Pittsburgh’s serious police-community divide through internal and external reform and a more data-driven community policing model, as well as boosting the morale of PBP’s 856 sworn officers.32 In McLay’s first year, PBP focused on expanding police presence at community events, implementing accountability 14 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS structures (including the Office of Professional Standards), inventorying PBP’s data and data systems, and diversifying the police force through changes in hiring practices.33 The PBP has an extensive history of reforming its problematic policing practices. As the first department investigated after DOJ acquired the power to launch pattern of practice lawsuits, the PBP entered a federal consent decree with DOJ in 1997 (Davis et al. 2002).34 The initial federal investigation found pervasive evidence of excessive uses of force, false arrests, improper searches and seizures, failure to discipline officers sufficiently, and failure to supervise officers, as well as uses of racial slurs by officers against African American residents (PERF 2013).35 From 1997 to 2002, the PBP implemented extensive reforms and policies consistent with the decree in areas such as reporting; investigation and audit processes; use of force, seizures, and traffic stops; required rotation;36 promotion and reassignment practices; officer discipline; community relationships; and training.37 In 2012, the ACLU filed a lawsuit against PBP alleging hiring discrimination that was resolved through a 2015 settlement requiring PBP to reform its hiring practices, establish a committee to oversee hiring reform, and compensate African American applicants rejected between 2008 and 2014.38 Several stakeholder interview respondents identified officer morale as one of the PBP’s main challenges when it began National Initiative activities. Some PBP officers felt that they were frequently under attack by city leadership. For example, Mayor Peduto publicly described PBP’s culture as mediocrity at best, and corruption at worst”; though he later apologized, his relationship with the department remained strained.39 When the National Initiative commenced in Pittsburgh, PBP officers had worked without a formal contract since January 2014 because of ongoing disagreement between city officials, the department, and the Fraternal Order of Police regarding officer pay, requirements, benefits, and other issues.40 There was also contentiousness regarding a residency requirement for officers, which an appellate court upheld in January 2016.41 Several major corporations and foundations in Pittsburgh have supported reform efforts: in March 2016, for example, Chief McLay announced that PBP would receive a $100,000 grant from the Heinz Endowments to fund trainings to improve police-community relations. Several local universities— including Carnegie Mellon, the University of Pittsburgh, and Duquesne—have also partnered with PBP in various ways. Pittsburgh is home to many civic engagement and community groups, and though many have voiced sentiments and led efforts critical of PBP, they could be resources for the department as it seeks to engage more positively with the community. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 15 Stockton, California Stockton is a city of 302,000 in north-central California with a 2014 per capita income of $19,927 and 25.8 percent of residents living below the poverty level.42 According to the most recent census,43 residents were predominantly Hispanic or Latinx (40.3 percent), non-Hispanic white (22.9 percent), or Asian (21.5 percent), with smaller populations of African American (12.2 percent) and multiracial (6.9 percent) residents. The 2008 housing market crash caused home values in Stockton to decrease 67 percent in five years and unemployment to rise to 17.2 percent. In 2010, it also ranked among the 10 cities with the highest violent crime rates nationwide (Logan and Stults 2011).44 The Stockton Police Department employed 409 sworn officers as of 2015, its highest staffing level since entering a period of a major financial decline in 2009, which was prompted by the city’s budget cuts and subsequent bankruptcy in 2012.45 That bankruptcy forced the SPD to lay off nearly one- quarter of its police force, helping drive the city’s unemployment rate above 20 percent.46 In response to these layoffs, SPD disassembled its narcotics force and cut back on community policing. Violence in Stockton reached new highs during this period.47 Eric Jones became SPD chief in March 2012 after six months of interim leadership following the previous chief’s resignation in 2011. In 2013, 32.9 percent of sworn SPD officers were officers of color, whereas three-quarters of Stockton residents were of color. SPD has since begun several initiatives to increase diversity and retention within the police force. In 2015, it began drawing funds from the newly implemented (and voter-approved) sales tax Measure B. Two-thirds of the tax revenues were intended to fund 120 additional officers over three years, and the remaining 35 percent was directed toward city efforts to recover from bankruptcy as well as services for residents, businesses, and property owners.48 Stakeholder interviews and site calls indicated that officer morale has improved under Chief Jones’s tenure and as the city and department have seen economic recovery. The SPD provided its officers eight hours of procedural justice training before joining the National Initiative, the most extensive exposure any of the participating police departments had to the concepts. Stockton is home to many civic engagement and community groups. These include the Community Advisory Committee, a diverse group that works closely with the department and chief and meets monthly to discuss problems faced by the various Stockton communities it represents. Other prominent groups include Agape Villages Inc., the Center for Community Involvement, El Concilio Council for the Spanish Speaking, Lao Family Community Empowerment, the San Joaquin Pride Center, the Women’s Center for Youth and Family Services of San Joaquin County, and a host of associations that represent Stockton’s diverse communities. 16 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS We have plenty of reason why we should be part of this initiative. The reasons are concentrated in our desire to live in peace, live with a sense of safety that has been so elus ive for our community for far too long. Stockton community resident Baseline Levels of Community Trust in the Police Soon after the National Initiative sites were selected, Urban conducted surveys of residents living in the communities with the highest levels of crime and concentrated disadvantage in each city. The results verified that the issues the National Initiative was designed to address were present and salient in all six sites. Residents in these areas perceived low levels of procedurally just treatment from the police and concerning levels of biased treatment (see La Vigne, Fontaine, and Dwivedi 2017). Relatively few people across all sites thought that police always or almost always engaged in such procedurally just behaviors as respecting people’s rights (30.2 percent), giving people a chance to tell their side of the story before deciding what to do (30.1 percent), treating people with dignity and respect (30.0 percent), and making fair and impartial decisions (26.0 percent). Moreover, substantial proportions of respondents agreed or strongly agreed that the police in their communities suspect people of criminal behavior because of their race or ethnicity (47.1 percent), act on personal prejudices or biases (51.4 percent), and treat people differently because of their race or ethnicity (55.5 percent). Urban reviewed the notes from the first wave of stakeholder interviews, which included police officers and residents and focused on issues cities were facing with police-community trust when they joined the National Initiative. The interview responses help explain this lack of trust. Although respondents raised a wide variety of issues, some common themes included the following: perceived inequality in effective protection and attentiveness to certain neighborhoods and communities, such that residents felt their neighborhoods or communities received less attention and protection than others issues with overenforcement or differential enforcement of lower-level offenses (such as traffic offenses) instead of more serious crimes lack of mutual understanding, meaningful connections, and engagement NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 17 the racial and ethnic composition of the police department differed from the communities they served, and in some sites most officers resided outside the city Having presented these contexts, this report turns to the specific interventions the National Initiative fielded to address these issues, starting with the training of police officers in the concepts of procedural justice and implicit bias. 18 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Chapter 2: National Initiative Trainings Training for police officers in the concepts of procedural justice and implicit bias was a foundational piece of the National Initiative implementation package, intended to promote trust-building interactions between officers and the public. The training component was the most resource-intensive intervention for the participating police departments, which collectively delivered nearly 100,000 person-hours of training. Officers received substantial exposure to the National Initiative concepts during these trainings, and training surveys and stakeholder interviews indicated widespread (though not universal) receptivity to the concepts. Innovations in training implementation included the creation of community-facing versions of the training to engage residents on the core procedural justice and implicit bias concepts. The National Initiative Training Continuum Several US police agencies implemented procedural justice and implicit bias trainings before the National Initiative launched, though there is limited research evidence regarding their effectiveness. The limited literature on procedural justice reports mixed but somewhat encouraging findings.49 Moreover, police trainings that educate officers on racial bias and cultural competence have become commonplace. At least 85 percent of training academies surveyed by the International Association of Directors of Law Enforcement Standards and Training50 reported increasing implicit bias trainings for new recruits since 2014, and 66 percent reported adding them for existing officers.51 Despite their proliferation , such trainings have not been systematically evaluated, and there is virtually no empirical evidence supporting their effectiveness in the domain of policing (Paluck and Green 2009). However, other fields have produced some promising evidence supporting the effectiveness of implicit bias training.52 The National Initiative procedural justice and implicit bias training consisted of the following three full-day (eight-hour) segments: Procedural Justice and Police Legitimacy (conceptual procedural justice), based on a curriculum developed by Yale Law School partners Tracey Meares and Tom Tyler in collaboration with the Chicago Police Department, provides a conceptual overview of NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 19 procedural justice. It focuses on how procedural justice enhances community cooperation with the police and supports effective policing, and it explores how officers can incorporate procedural justice principles into their daily activities. The first segment also discusses policing from a historical perspective in terms of its relationship with communities of color.53 A Tactical Mindset: Procedural Justice and Police Legitimacy (tactical procedural justice) was also based on a curriculum Meares and Tyler developed. A more tactical approach to procedural justice, it uses simulations and scenario-based exercises and teaches techniques for applying procedural justice on the job.54 Tactical Perception: The Science of Justice (implicit bias) was based on the work of Center for Policing Equity President Phillip Atiba Goff. Goff developed the curriculum in collaboration with Kimberly Burke (also of CPE) and Al Ferreira of the CPD training division. It engages law enforcement officers to think critically about racial bias using Goff’s framework, called “identity traps.” Identity traps are situations that make people more likely to allow psychological factors such as implicit bias and threats to one’s self-concept) to facilitate behaviors inconsistent with one’s values.55 There are two kinds of identity traps: fast and slow. Fast traps, or implicit biases, are attitudes and stereotypes not consciously accessible through introspection. Slow traps occur when a person’s identity is salient or important in a particular situation, that person’s ability to manage others’ perceptions of them are low, and failure to validate the person’s identity could lead to behaviors with negative consequences. Individuals are more likely to be subject to both fast and slow traps when they are mentally taxed, in a bad mood, feeling threatened, a novice, making quick decisions, or multitasking. Each training consisted of four to six modules, outlined in table 2.1. 20 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS TABLE 2.1 National Initiative Training Modules CPJ TPJ IB Module 1 Introduction to the concepts of procedural justice and legitimacy, including the benefits of legitimacy to police. Conceptual review of procedural justice. Introduction, how implicit bias operates (fast and slow traps), connection to procedural justice trainings. Module 2 How law enforcement work can lead to cynicism; cynicism’s effect on professional performance and officer wellness; solutions for addressing cynicism. Transparency as a way to develop trust (scenario 1). Exploration of fast traps and how they influence decisionmaking. Module 3 Detail on procedural justice and its four components: voice, neutrality, respect, and trustworthiness. Community perceptions of the police; informing the community about procedural justice; considering its role in that process (scenario 2). Exploration of slow traps and how they influence decisionmaking. Module 4 History of policing and race, and relations to marginalized communities. Treatment matters (scenario 3). Strategies for defusing traps to improve decisionmaking. Module 5 Positive examples of procedural justice in practice; training is wrapped up. Communication matters/tactical communication (scenario 4). N/A Module 6 N/A The higher purpose of policing. N/A Implementation Experiences and Fidelity Between December 2015 and April 2018, the National Initiative police departments delivered the full three-part training curriculum to all sworn officers. In this section, we describe how those trainings unfolded in the field as well as barriers to and facilitators of successful implementation. We also describe innovations that emerged, particularly police departments’ adaptation of the curricula to address procedural justice internally and the development of trainings to teach community members the core procedural justice and implicit bias concepts. Preparing for the Trainings The participating police departments began the training preparation process by identifying training teams. Following the National Initiative partners’ advice, police leadership selected training teams composed of officers with the experience and standing to maximize credibility when delivering the procedural justice and implicit bias material. The six training teams travelled to Chicago in October 2015 to participate in a train-the-trainer workshop.56 During this workshop, CPD instructors prepared NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 21 the trainers to deliver the two-part procedural justice and police legitimacy curriculum (CPJ and TPJ), which had been administered to more than 11,000 CPD officers over the previous four years. The training teams also received a preview of the material on identity traps and a draft version of the implicit bias curriculum. Upon returning to their departments, the PJ training teams began preparing the training for local delivery. Each team divided the modules among the trainers, practiced delivering them, and tailored the curriculum by incorporating images and videos from their own communities and departments into training slide decks. As the training teams delivered CPJ and TPJ,57 CPE developed the implicit bias curriculum iteratively by soliciting feedback from the training teams as they refined the curriculum. According to interviews with trainers, multiple sites appreciated CPE’s receptivity to that feedback. Trainers generally felt that the resulting IB curriculum was high quality, and some called it the best of the three training components. At the same time, bias is a loaded idea in policing because officers are sensitive to accusations of racism, and many trainers found it intimidating to prepare for. As a Pittsburgh PJ trainer put it, “[implicit bias] has potential to be the bomb that nukes it all, because it’s such a difficult topic and it’s so easy to get wrong as instructors.” A Stockton trainer echoed this sentiment: “It’s huge, and they only have one opportunity to do it. If they screw it up, or alienate officers, it will be difficult to recover.” Because implicit bias was a new curriculum, CPE included substantial piloting and peer exchanges. Every procedural justice training team from a National Initiative site participated in at least three peer exchanges related to launching the IB training, and five of the six sites hosted an exchange. In the last of a series of peer exchanges before the IB trainings began, FWPD hosted trainers from CPD and each of the National Initiative sites in June 2017. National Network for Safe Communities created remote peer consultation structures that played important roles in preparing departments for TPJ and IB and in refining and improving their delivery of the trainings. These included an email listserv, launched in April 2016 for the CPD trainers and the six department training teams, that allowed them to ask questions, troubleshoot training strategies, and share success stories. The teams communicated via the listserv through August 2018. In April 2017, NNSC and CPE began monthly calls with all six training teams to discuss issues and challenges. Training Delivery All six sites delivered the three components of the training continuum to all sworn officers in their police departments, with consistency in duration and content. The trainings were delivered to cohorts with officers from mixed ranks and across department divisions, functions, and geographical areas. 22 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Including higher-ranking officers in the trainings had important implications for discussions about internal department issues. Figure 2.1 shows the sites’ training timelines. Every site launched its first training within four months of the October 2015 train-the-trainers workshop in Chicago (Pittsburgh started first in December 2015). There were two significant variations in the training delivery. First, Birmingham PD delivered CPJ and TPJ at a two-day, 16-hour combined training, whereas the other sites had their officers complete CPJ before anyone began TPJ. Second, Stockton PD delivered the CPJ curriculum before joining the National Initiative, so its first National Initiative training was TPJ. By July 2016, all officers across the six departments had participated in the CPJ training (with minor exceptions such as officers out on extended leave); by the end of 2016, all the sites had completed TPJ, except for FWPD, which took longer because of its size (table 2.2). Five departments delivered all 24 hours of training to all sworn officers by July 2017, and Fort Worth completed the trainings in April 2018. FIGURE 2.1 National Initiative Training Calendar CPJ TPJ CPJ & TPJ IB 2015 2016 2017 2018 D J F M A M J J A S O N D J F M A M J J A S O N D J F M A Birmingham Fort Worth Gary Minneapolis Pittsburgh Stockton Officer Responses to the Trainings CHANGES IN OFFICER AGREEMENT WITH TRAINING CONCEPTS As part of its evaluation, Urban assessed the trainings to determine whether participating officers increased their understanding of and agreement with the core concepts. Learning assessment surveys captured officer views on core concepts of procedural justice and implicit bias before and after the training (see table 2.3). As such, survey results cannot speak to how lasting any changes in attitudes and knowledge were or how they may have manifested in officer behavior. Although the conclusions that may be drawn from the assessments are limited, we included them in the evaluation because changes in knowledge and attitudes are important goals of the trainings. Changes in knowledge and attitudes are a precursor to changes in behavior and a necessary foundation for practice changes. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 23 TABLE 2.2 Training Learning Assessment Surveys Completed Birmingham Fort Worth Gary Minneapolis Pittsburgh Stockton CPJ pretest 818 1,502 220 864 405 CPJ posttest 1,459 225 895 630 TPJ pretesta 442 TPJ posttest 804b 1,602 0c 947 0c 479 IB pretest 571 1,711 88 815 526 140 IB posttest 581 1,592 86 814 561 134 a TPJ pretest administered in Stockton only, as they had delivered CPJ before joining the National Initiative. b Birmingham delivered CPJ and TPJ as a two-day training block; posttest administered at the conclusion of the second day. c TPJ posttest assessments were omitted in Gary and Pittsburgh. Urban conducted one-sample t-tests on individual items and the procedural justice scale for differences between the pre-CPJ baseline and post-CPJ and post-TPJ responses. We observed statistically significant increases in agreement with the procedural justice principles from the pretest to the posttest (see table 2.3). In general, the increases were evident by the conclusion of CPJ and maintained through the end of TPJ. It is also worth noting officers’ generally high levels of agreement with procedural justice principles before the CPJ training. Principles related to trusting the public received the lowest levels of agreement, which is consistent with the evaluation of the Chicago Police Department’s procedural justice training (Skogan, Van Craen, and Hennessy 2015). We observed increases in levels of agreement with procedural justice principles in all six National Initiative sites, with modest variation in the initial levels of agreement and scale of change (see figure 2.2). 24 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS TABLE 2.3 Mean Differences on Procedural Justice Items Pre-CPJ Post-CPJ Post-TPJ Officers should at all times treat people they encounter with dignity and respect. 3.59 3.69** 3.73** It is important to give everyone a good reason why we are stopping them. 3.39 3.54** 3.55 Listening and talking to people is a good way to take charge of situations. 3.41 3.60** 3.63* Police have enough trust in the public for them to work together effectively. 2.46 2.84** 2.91** If people ask why we are treating them as we are, we should explain our actions as soon as we can. 3.20 3.47** 3.56** When dealing with citizens' concerns, officers need to explain what will happen next when they are done at the scene. 3.27 3.49** 3.50 People should be treated with respect, regardless of their attitude. 2.97 3.24** 3.28 Officers need to show an honest interest in what people have to say, even if it is not going to do anything. 3.23 3.46** 3.49 Officers should treat citizens as if they can be trusted to do the right thing. 2.93 3.20** 3.25** It is important that we understand, follow, and protect the rights of the people with whom we come into contact.. 3.62 3.70** 3.76** It is very important that officers appear neutral in the application of legal rules. 3.58 3.73** 3.74 Total procedural justice scale 3.24 3.46** 3.49** Notes: Valid N = 11,056. Cronbach’s alpha = 0.9. Response options ranged from 1 (strongly disagree) to 4 (strongly agree). Statistically significant differences are indicated for post-CPJ relative to pre-CPJ, and post-TPJ relative to post-CPJ and were assessed using t-tests; *p < 0.05, **p < 0.01. Because Stockton’s police department conducted a CPJ training before participating in the National Initiative, preassessment surveys were not fielded in that site. Additionally, postassessment surveys for the TPJ training were not fielded in Gary or Pittsburgh. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 25 FIGURE 2.2 Changes in Agreement with Procedural Justice by Site Notes: Valid N = 11,056. Cronbach’s alpha = 0.9. Response options ranged from 1 (strongly disagree) to 4 (strongly agree). Statistically significant differences are indicated for post-CPJ relative to pre-CPJ, and post-TPJ relative to post-CPJ and were assessed using t-tests; *p < 0.05, **p < 0.01. Because Stockton’s police department conducted a CPJ training before participating in the National Initiative, preassessment surveys were not fielded in that site. Additionally, postassessment surveys for the TPJ training were not fielded in Gary or Pittsburgh. BOX 1 Internal Procedural Justice Officers across the National Initiative sites raised concerns about whether police supervisors (and their departments writ large) would apply procedural justice concepts internally. Trainers in several sites described incorporating discussion of internal procedural justice into the conceptual procedural justice training, often alongside the discussion on cynicism in module 2. Pittsburgh received such strong demand for addressing procedural justice internally that the PBP trainers added a module on applying internal procedural justice concepts during CPJ implementation. Supervisors were present during most trainings, providing officers of all levels an opportunity to voice concerns about internal issues. Several trainers believed that the success of the procedural justice trainings ultimately depended on improving internal procedural justice. As one PBP trainer said, “I think the class was sort of temporarily successful in the sense that if the administration does not follow 3.33 3.28 3.29 3.17 3.05 3.58** 3.52**3.52** 3.43** 3.19** 3.41 3.58 3.45**3.45 3.63** Birmingham Fort Worth Gary Minneapolis Pittsburgh Stockton Pre-CPJ Post-CPJ Post-TPJ 26 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS through on making [internal procedural justice] a critical part of every interaction, then we will have lost any ground we gained with that class.” Findings from the learning assessment showed that 17 percent of PBP officers who commented on the training in an open-ended question said that follow-up and accountability from department management on internal issues discussed during training was critically important. The implicit bias learning assessment surveys measured officers’ views on how bias operates and influences officer conduct. Findings for all National Initiative departments are shown in figure 2.3 and by site in table 2.4. For most items, officers’ views were more aligned with the concepts after the training than before. This was particularly true for items about the influence of stereotypes on individual behavior and the need to mitigate them through reflection and awareness. For instance, officers were more likely to agree after the training that stereotypes could influence their interactions with particular groups without their awareness and that even well-meaning people have biases, and they were less likely to agree that nothing could be done about implicit biases if they are unconscious. However, agreement that biases affected their behavior was essentially unchanged. FIGURE 2.3 Officer Agreement with Implicit Bias Concepts Notes: Valid N = 7,428. Response options ranged from 1 (strongly disagree) to 5 (strongly agree). Statistically significant differences were assessed using t-tests. *p < 0.05, **p < 0.01. 3.22 3.78 2.49 3.24 2.72 2.73 3.34 3.78 3.19 4.04** 2.28** 3.30* 2.79* 3.02** 3.58** 4.09** My biases do not affect my behavior Even well-meaning people have biases If implicit biases are unconscious, then there's nothing I can do about them I am worried that someone will prejudge me because I am a police officer My concern about being prejudged as a police officer could impact my behavior Stereotypes about particular groups could influence my interactions with them, without my awareness At times, situations can influence our decisions and behaviors, more than personal character Being reflective and aware of how I respond to people and situations can increase my effectiveness as a police officer Pretraining Posttraining NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 27 Because the aggregate findings could mask significant variation across the sites, we also examined responses by department. As shown in table 2.4, Birmingham officers’ views differed from those of other sites’ officers before the IB training : BPD officers agreed less with the statements, “my biases do not affect my behavior,” “even well-meaning people have biases,” and, “being reflective and aware of how they respond to people and situations can increase their effectiveness as a police officer,” and they agreed more that “nothing can be done about biases if they are unconscious.” Following the training, BPD officers’ views were more aligned with the other officers’ views, signifying a larger change in views in Birmingham than in any other site. TABLE 2.4 Officer Agreement with Implicit Bias Concepts, by Department Pretraining Posttraining B FW G M P S B FW G M P S My biases do not affect my behavior 2.46 3.41 3.51 3.07 3.52 3.50 3.20** 3.21** 3.20 3.06 3.28** 3.29 Even well-meaning people have biases 2.39 4.19 4.00 3.63 4.04 4.07 3.93** 4.05** 3.91 4.07** 4.02 4.40** If implicit biases are unconscious, there is nothing I can do about them 3.25 2.40 2.57 2.20 2.50 2.21 2.72** 2.15** 2.00** 2.24 2.37* 2.00 Stereotypes about particular groups could influence my interactions with them, without my awareness 3.22 2.64 2.86 2.63 2.59 2.78 3.26 2.86** 3.03 3.17** 2.93** 3.20** Being reflective and aware of how I respond to people and situations can increase my effectiveness as a police officer 2.17 4.25 4.00 3.59 4.11 4.32 4.17** 4.07** 3.91 4.00** 4.09 4.53* Notes: B = Birmingham, FW = Fort Worth, G = Gary, M = Minneapolis, P = Pittsburgh, S = Stockton. Valid N = 7,428. Statistically significant differences from pretraining to posttraining were assessed using t-tests. *p < 0.05, **p < 0.01. Response options ranged from 1 (strongly disagree) to 5 (strongly agree). OFFICER VIEWS ON TRAINING AND TRAINER QUALITY The learning assessments found that officers across departments viewed the trainings positively. Officers were asked questions about training quality and delivery after all three trainings, allowing for comparison of officer perceptions of each component. Officers rated the three trainings favorably (see figure 2.4). The implicit bias training received the highest overall rating, and its instructors were the most highly rated. Ratings for CPJ and TPJ were similar. 28 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Interview respondents remarked that strategically selecting particular officers to serve as trainers was fundamental to successful trainings. This was supported by the high ratings those trainers received and by officers’ responses to the open-ended survey gauging the most valuable aspects of the training. Officers consistently commended instructor quality in survey responses (see table 2.5). Praise for instructors commonly included terms such as “sincerity,” “commitment,” and “credibility.” FIGURE 2.4 Overall Training and Instructor Ratings Notes: Valid N = 8,011. Response options ranged from 1 (unsatisfactory) to 5 (excellent). Statistically significant differences are indicated for post-TPJ relative to post-CPJ, and post-IB relative to post-TPJ, and were assessed using t-tests; *p < 0.05, **p < 0.01. Officer views of the training material’s applicability were generally favorable, but the relative ratings differed from those for overall quality and instructor quality (figure 2.5). On average, officers considered the implicit bias material the least applicable and relevant, and the least likely to benefit other officers and the department. Scores for the applicability and relevance of CPJ and TPJ were similar. The ratings of the IB trainings most similar to the other two concerned learning new information and skills. 3.86 3.91 3.92 3.92 3.74 3.80 3.86 3.87 3.85 3.60 4.40 4.48 4.52 4.49 3.92 How would you rate the instructor(s) on how they used relevant examples? How would you rate the instructor(s) on how they responded to questions? How would you rate the instructor(s) on how they knew the subject matter? How would you rate the instructor on how they used audience participation How would you rate the training overall? Post-CPJ Post-TPJ Post-IB NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 29 FIGURE 2.5 Views on Training Applicability Notes: Response options ranged from 1 (strongly disagree) to 4 (strongly agree). Statistically significant differences are indicated for post-TPJ relative to post-CPJ, and post-IB relative to post-TPJ, and were assessed using t-tests; *p < 0.05, **p < 0.01. Data from this figure’s last two items did not include responses from Fort Worth. The stakeholder interviews generally echoed these findings about training performance. Multiple officers from all sites and across the 2016, 2017, and 2018 interview waves described the trainings as highly successful. (Only three community respondents expressed an opinion about the impact of the trainings, two of whom believed they were successful.) A Minneapolis PJ trainer said of CPJ, “They really liked it, I was really surprised. There was very minimal pushback, especially from the veteran officers who I was expecting to give a lot of pushback. Surprisingly, it was the newer officers who had more issues.” Trainers from two cities said the curriculum received the best participant ratings they had ever gotten for trainings. However, some trainers said officers were buying in at different degrees. As a Pittsburgh PJ trainer put it, “I think it’s probably 10 to 15 percent who are completely on board; 50 percent who say I’m kind of on board with it, but we have calls, I’ll do it when I can; and the rest won’t buy into it.” Trainers in three departments reported that a “vocal minority” of officers responded negatively to the CPJ training. Trainers from multiple sites said officers were skeptical at the beginning of CPJ, but left feeling supportive of and/or pleasantly surprised by it. A PJ trainer in Fort Worth said, “I think people came into that training wanting to hate the National Initiative and basically very skeptical of what we were doing, 3.44 3.16 3.10 3.20 3.47 3.35 3.39 3.37 3.38 3.05 3.01 3.17 3.44 3.30 3.28 3.30 3.27** 3.17** 2.95** 2.98** 3.15** 3.01** 3.01** 3.05** The training met my expectations for learning I learned new information from the training that will help me in my job I learned new strategies and skills from the training that will help me in my job The training will help me perform my job more effectively The training was relevant to my job duties I expect to apply much of what I learned from this training to my work I would recommend this training for other police officers My organization will benefit from having completed this training Post-CPJ Post-TPJ Post-IB 30 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS and I believe that we had a lot of people changing from skeptics because of the training.” Interview ees also indicated that skepticism waned as training implementation proceeded and positive attitudes toward the trainings spread through the department. Some PJ trainers shared that skepticism when first exposed to the trainings. A Pittsburgh PJ trainer said, “When [Chief] McLay asked me to do this initially, I was like I don’t know if I buy into this. But through using this in my daily work, I see really how much the community appreciates this and how much they care … And because I buy into that, I try to impress that on the younger officers and explain to them, if you go out and do what I did for 15 years and just try to bring everyone to jail, you’re just going to make it hard on yourself.” According to trainers, some officers found TPJ redundant, while others said its focus on application made it better than CPJ. Minneapolis trainers, for example, believed that officers liked TPJ better than CPJ because it involved more discussion and fewer lectures. The training learning assessment data suggest officers may not have seen much difference in quality, although in the open response questions about suggestions for improvement, Fort Worth officers were particularly likely to describe CPJ as redundant (11 percent of respondents). As with CPJ, trainers in several cities said they had to overcome officers’ initial skepticism about the implicit bias training, but that officers tended to leave with a positive impression of it. A Pittsburgh PJ trainer gave a detailed description of how IB was received and the experience of teaching it that illustrates several common themes from interviews of trainers: I actually thought [Implicit Bias] would be the worst to teach, but it’s my favorite. Because when you think about it from the officers’ perspective … everyone knows [conceptual procedural justice]. They might not know the language or practice it all the time, but [CPJ] is really just ‘don’t be an asshole, treat people with respect.’ And the pushback we got on [CPJ] was mainly ‘this is common sense, why do I have to do this.’ [Tactical procedural justice], cops love tactics. They love to learn about good tactics and how to end a situation tactically and whatnot, and the practice. The officers enjoy that kind of stuff, so I didn’t see that as being a tough class to teach. But implicit bias], in my mind I thought it was going to be really brutal, because in their minds you’re teaching them not to be racist. But [our trainer] … did a phenomenal job, and really helped construct the class in a way so instructors could present it in a way that defused people. It didn’t point fingers or accuse people of being racist. He put in examples that weren’t terribly serious but just got people to understand implicit bias. So I enjoyed teaching the class a lot, but also found it fascinating because it’s all about human behavior. Like [CPJ], you know how you’re supposed to treat people, but nobody really knows about implicit bias. MODE OF DELIVERY Across all trainings, survey responses suggest that the officers appreciated the group discussions and interaction, including the opportunity to hear perspectives and discuss issues of policing with other officers. They also noted the importance of having a space where they could discuss issues they faced in NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 31 policing, including internal department issues, without judgment or fear of reprisal. Officers in all six departments considered the open discussions and opportunities to exchange views with other officers among the most valuable parts of the training continuum (see table 2.5). Officers credited these discussions and spaces to the instructors’ skill and experience, which is reflected in the high ratings of instructor quality and examples citing instructors among the most valued parts of the trainings. Group discussions also tied to material on officer wellness and stress. One Gary trainer discussed how these elements, by design, mitigated the cynicism that policing can create and reminded participants why they became officers: I think it allows us or reminds us … we were all just wide-eyed young people when we came to this agency. And, so, I think the training when it's presented in a particular way takes us back to that to all the hope and you know, all of the excitement we had for the job before we became so disillusioned and frustrated and started seeing everybody through the same Blue Lens. Everybody's horrible. But this training takes us back to when we were like fresh and young and excited and wanted to help. And I always say when it is presented a certain way; so, when it's presented in that way you can have it almost energizes officers. 32 NATIONAL INITIATIVE IMPLE MENTATION ASSESSMENT FINDINGS TABLE 2.5 Most Valued Training Components Most common open responses from training assessment surveys CPJ TPJ IB Birmingham CPJ & TPJ n = 538 IB n = 303 open discussion (13.2%) focus on treating people with respect (12.8%) focus on giving voice/listening (8.6%) all/everything (8.0%) focus on communication/relating to people (6.3%) instructor quality (5.9%) four elements of procedural justice (5.6%) conceptualization of legitimacy (5.2%) assisted with self-awareness and bias mitigation (15.5%) all/everything (11.9%) discussion of universality of bias 10.2%) videos (8.6%) instructor quality (7.6%) examples from policing experience 6.9%) psychology of implicit bias (6.9%) open discussion (5.6%) traps concepts (5.3%) Fort Worth CPJ n = 681 TPJ n = 778 IB n = 601 all/everything (18.4%) refresher/reaffirmation of good policing 8.2%) examples from policing experience 8.2%) instructor quality (6.0%) videos (5.9%) open discussion (5.6%) videos (15.2%) open discussion (14.9%) all/everything (12.0%) instructor quality (6.6%) focus on communication/relating to people (6.0%) videos (20.0%) open discussion (13.6%) all/everything (10.3%) instructor quality (7.8%) psychology of implicit bias (7.8%) assisted with self-awareness and bias mitigation (6.0%) discussion of universality of bias 5.2%) Gary CPJ n = 126 IB n = 47 open discussion (22.2%) videos (13.5%) all/everything (11.9%) focus on communication/relating to people (9.5%) all/everything (9.3%) assisted with self-awareness and bias mitigation (9.3%) traps concepts (7.0%) discussion of universality of bias 5.8%) NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS 33 CPJ TPJ IB Minneapolis CPJ n = 441 TPJ n = 416 IB n = 266 discussion of race and history of policing 26.1%) instructor quality (25.2%) videos (10.4%) all/everything (8.2%) examples from policing experience 7.0%) open discussion (6.3%) refresher/reaffirmation of good policing 5.9%) videos (21.6%) instructor quality (15.9%) open discussion (13.9%) refresher/reaffirmation of good policing (7.2%) group work (5.0%) instructor quality (18.4%) videos (16.2%) open discussion (15.0%) examples from policing experience 13.5%) traps concepts (7.9%) assisted with self-awareness and bias mitigation (6.4%) Pittsburgh CPJ n = 264 IB n = 285 open discussion (23.9%) instructor quality (13.6%) videos (11.0%) engagement with internal issues/internal procedural justice (8.3%) examples from policing experience 5.7%) assisted with self-awareness and bias mitigation (11.6%) discussion of universality of bias 10.9%) examples from policing experience 10.5%) open discussion (9.1%) traps concepts (8.4%) instructor quality (7.7%) videos (7.4%) Stockton TPJ n = 210 IB n = 84 scenarios/role plays (22.4%) open discussion (15.7%) examples from policing experience (14.3%) videos (12.9%) citizen views of police (7.6%) focus on treating people with respect (6.2%) all/everything (5.7%) focus on communication/ relating to people (5.7%) refresher/reaffirmation of good policing (5.7%) open discussion (27.4%) assisted with self-awareness and bias mitigation (11.9%) videos (10.7%) discussion of universality of bias 8.3%) examples from policing experience 7.1%) psychology of implicit bias (6.0%) Source: Urban Institute analysis of open text responses to National Initiative training assessment surveys. Notes: Reported percentages are of valid responses to open text questions. Training surveys were not administered for TPJ in Gary or Pittsburgh. 34 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Community-Facing Training Over time, departments showed interest in doing parallel procedural justice and implicit bias trainings for their communities. Officers noted that the trainings were focused on the police’s obligations to the community and asked whether there would be parallel interventions regarding the community’s obligations when engaging with police. Trainers noted in interviews that community groups had also expressed interest in receiving the trainings. The National Initiative technical assistance team helped the police departments prepare a community-facing version of the procedural justice training as they concluded the CPJ and TPJ trainings. CPE developed an abbreviated curriculum for the community- facing IB training while preparing it for the police departments. Some interview respondents expressed reservations about using the term “training” to describe this work with the community because community members were not receiving training analogous to that given to officers. Several departments were intentional about messaging with communit ies. Interview respondents described three main goals for the community-facing versions of the trainings: 1) to inform the community of the concepts officers were learning through the National Initiative, so residents would better understand what the police were doing internally; (2) to elicit openness and responsiveness from the community when they recognized procedurally just policing; and (3) to help communities understand their role in fostering positive police-community relationships. COMMUNITY TRAINING CONTENT AND DELIVERY Pittsburgh was the first site to begin developing a community-facing version of the procedural justice trainings. Their training team developed a 2- to 3-hour version of the CPJ curriculum to share with community members. Fort Worth adapted similar sessions intended to engage youth. The curriculum for these sessions covered the relationship between procedural justice, legitimacy, and crime control, the four elements of procedural justice and how communities should expect police to deliver them, and the historical perspective on policing officers received. It also included a preview of the implicit bias training. Pittsburgh delivered the first community PJ training on June 29, 2016, at a summit hosted by the University of Pittsburgh School of Social Work, and by September 2017 it had been delivered to nearly 20 groups in Pittsburgh (those groups had generally requested it). Community PJ trainings or similar sessions were rolled out in all the National Initiative sites in 2016 and 2017. Center for Policing Equity developed a community-facing implicit bias curriculum designed to foster dialogue between police and communities. The training had four goals: (1) to understand the psychological science of bias and its influence on behaviors in universal and policing context s, (2) to NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 35 create a shared language to bridge the divide between law enforcement and the communities they serve, (3) to engage in dialogues that build empathy and increase self-reflection, and (4) to empower participants with tools for improving their understanding of and responses to situations vulnerable to bias. Community members cofacilitated sessions on the IB content in all six sites. Center for Policing Equity staff trained Birmingham, Gary, and Minneapolis trainers in summer and fall 2017, and Fort Worth, Pittsburgh, and Stockton trainers in early 2018. Participants in some sites incorporated core content from the training curricula into targeted community outreach and engagement efforts. For example, Fort Worth PJ trainers distilled takeaways from the procedural justice curriculum into a youth-facing training guide that school resource officers used to lead trust-building conversations (referred to as “Youth Engagements and Conversations”) with students (conversations thus incorporated elements of both the procedural justice and reconciliation pillars). (For more details, see NNSC [2018b], a case study on PJ training and mentorship training for school resource officers in Fort Worth.) In Birmingham, the IB community curriculum was delivered to students at an alternative high school as part of three consecutive weeks of trust-building sessions that were similar in many ways to the reconciliation listening sessions (see chapter 3). Sustaining and Institutionalizing the Training Concepts Procedural Justice trainers, officers, and community members wanted to know what would come after the training, and the sites implemented several measures to sustain the training’s operational and structural impact. Refresher trainings were one such measure. Fort Worth planned to add a PJ refresher to its annual in-service training, and Minneapolis instituted an annual 4-hour refresher for all staff. One Minneapolis trainer said, “Every year we do use of force. Every year officers have to qualify with their handguns. And they really, when you think about it, the biggest, the best equipment they use the most is their mouth.” The MPD also trained all civilian staff on conceptual procedural justice and delivered a modified version of tactical procedural justice to further integrate the concepts within the department. Some departments formalized their training teams and tasked them with creating procedural justice units to advance procedural justice concepts throughout the department (see chapter 4). The training teams in the six sites trained some of their peer agencies to diffuse the procedural justice and implicit bias concepts in their regions. Interview responses indicated that expanding the trainings regionally was a promising approach to building local peer networks and embedding the concepts into policing and justice system practice. For instance, the Chicago PD trained would-be CPJ 36 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS and TPJ trainers from the Gary PD and seven of its neighboring jurisdictions. The Gary PJ training team then helped roll out procedural justice courses at the Northwest Indiana Law Enforcement Academy, which provides training for GPD and other regional law enforcement agencies. Gary also trained officers in the Lake County Sheriff’s Department, which shares policing responsibilities in Gary, most notably with the joint Metro Homicide Unit. Birmingham trained staff working in jails, Minneapolis and Birmingham trained their 911 dispatchers, and Pittsburgh trained 50 Port Authority Police Department officers. As of June 2017, Stockton PD had trained staff of the San Joaquin County Probation Department and District Attorney’s Office in procedural justice and delivered procedural justice training to more than 100 California agencies via train-the-trainers sessions. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 37 Chapter 3: Reconciliation Although reconciliation was one of the National Initiative’s three organizing pillars, the initiative did not have a formalized reconciliation process when it commenced. Of the three pillars, reconciliation was the most nascent and previously untested. National Network for Safe Communities developed a framework for a police-community reconciliation process, something that had not existed in the US. Though all six sites implemented reconciliation, they made differing degrees of progress implementing components of the framework. The process provided residents a space to raise issues and concerns that directly prompted changes in police practice. This section describes the reconciliation process the National Initiative developed, and how the sites implemented its components. The National Initiative Reconciliation Process The National Initiative’s reconciliation efforts focused particularly on communities of color and communities experiencing violence, “overpolicing” (intensive enforcement of low-level crimes and violations), and “underpolicing” (a failure to adequately prevent or solve serious violent crimes). National Network for Safe Communities developed several iterations of a reconciliation process during the National Initiative. It consulted with the National Initiative partner organizations, sites, and external advisors, most notably Susan Glisson of the William Winter Institute for Racial Reconciliation and later of Sustainable Equity, LLC), Priscilla Hayner of the Truth and Reconciliation Commission of South Africa (see, e.g., Hayner 2011), Fanie Du Toit, then-director of the Institute for Justice and Reconciliation in South Africa, and Lord John Alderdice of the Center for the Resolution of Intractable Conflict at Oxford. The reconciliation process framework as of the end of 2018 is shown in figure 3.1. Its purpose was to build and strengthen police-community relationships by facilitating the following: addressing differences in perspectives acknowledging past harms learning about and acknowledging past and present conflicts and harms facilitating sustained communication to bridge differences and identify common ground and goals explicitly and tangibly committing to changing practices and policies 38 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS FIGURE 3.1 National Initiative Reconciliation Framework The reconciliation framework consists of the following five key components: Fact-finding around harms the police have done in specific communities, from enforcing Jim Crow laws to current policies, practices, and dynamics with detrimental effects on safety and justice. Identifying harms and their impacts on diverse local and national communities must be done with care and intentionality. Results of the historical fact-finding are disseminated through department-wide procedural justice trainings and inform police leadership’s public acknowledgements of harm. Fact-finding on current problems provides departments a basis for identifying changes and reparative actions to better align them with community priorities and build community confidence in the reconciliation process. Acknowledgement of harm by police leadership to recognize the police’s past and present harms as well as ongoing problems that fuel mistrust between the police and community. Sincerely and specifically acknowledging harms launches the public-facing reconciliation process. Harms include officer-level issues (e.g., engaging in disrespectful treatment) and department-level issues (e.g., overuse of stop and frisk tactics). Sustained listening via listening sessions, where police provide community members an intimate, nonadversarial forum to share their experiences and insights about law enforcement. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 39 These sessions occur in small settings (generally between 8 and 20 participants) and are a key mechanism for identifying narratives and informing changes to policy and practice, which are reported back to community members in subsequent listening sessions. Narrative collection and sharing to capture the community’s common perceptions of police and the police’s common perceptions of their communities. Some narratives may illuminate common ground for joint action, such as shared commitment to reducing violence; others may highlight impediments to reconciliation, such as perceptions that communities are too tolerant of crime and violence or perceptions that the police deliberately engage in harmful actions. The trauma-informed collection of these narratives provides a record of lived experiences, local voice, and present harm. Sharing these narratives with the public is intended to facilitate mutual understanding and build empathy among more people than are able to participate in listening sessions. Explicit commitments to changing policy and practice in areas identified during listening sessions, in the form of concrete, reparative, and substantive modifications. Effective implementation of these changes is ensured through training and by holding identified parties accountable. Departments communicate changes to reconciliation process participants and the public. This is necessary to create confidence that the reconciliation process can effect meaningful change. BOX 2 Unique Subpopulation Interventions As an integral component of efforts to build police-community trust, the National Initiative included targeted outreach to groups with historically troubled and/or distrustful relationships with law enforcement. Such groups may experience dynamics in interactions with police that require unique outreach and interventions. These subpopulations (identified by the DOJ) included youth, the LGBTQIA+ community, Native Americans, immigrants (particularly Latinx immigrants), and victims of domestic violence and sexual assault. Some sites focused on additional populations, such as East African immigrants in Minneapolis and Pittsburgh. Interventions for each subpopulation were developed through the following nine steps: building a matrix of local resources and programs serving the subpopulation facilitating local partnerships across government agencies, advocacy and social service organizations, and community leaders to improve coordination and wraparound services launching reconciliation processes between the subpopulation and police 40 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS developing tools, guides, and manuals to help law enforcement serve the subpopulation creating and promoting communications that highlight efforts to address the subpopulation facilitating peer learning opportunities between National Initiative sites on how to best serve the subpopulation connecting sites with external training opportunities and resources to help them better serve the subpopulation supporting existing and emerging violence interventions that address the subpopulation providing ongoing technical assistance to police departments on existing efforts and projects pertaining to the subpopulation As the subpopulation intervention plan and reconciliation process progressed, and because many of the nine steps incorporate elements of reconciliation components (particularly around fact-finding and listening sessions), NNSC increasingly integrated them. Each site identified and incorporated at least one subpopulation in their reconciliation process, and those subpopulations became a priority of that work. These efforts are highlighted in the implementation discussion that follows. Reconciliation Process Implementation Across sites, reconciliation processes varied considerably in timing and scope. Some launched a process consistent with the framework soon after it was developed and had engaged in extensive reconciliation activity by the end of 2018. Others took longer, and some had made little progress by the end of 2018. However, all six sites implemented each of the reconciliation process components to some degree. Fact-Finding Fact-finding, which informed the reconciliation process in each site, was decentralized and occurred during the initial stages of the National Initiative, when the reconciliation framework was still being developed. Although focus groups and other community meetings during the first site visits were forms of fact-finding, the lack of documentation of those sessions and turnover in the National Initiative team that participated in those visits limited their utility for this purpose. The procedural justice training teams did important fact-finding work, investigating local historical contexts for police-community tensions and incorporating findings in the training module that covered the history of policing and race. For example, the Minneapolis PD worked with Native American community leaders to create a historical module about the Native American experience with policing in Minneapolis to be NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 41 incorporated in the procedural justice trainings. Urban’s baseline community surveys, though administered as evaluations rather than reconciliation support activities, provided insights into communities’ perceptions of police legitimacy and trustworthiness and their willingness to engage in crime control efforts (La Vigne, Fontaine, and Dwivedi 2017). Several departments incorporated these results into their acknowledgements of harm and procedural justice trainings. As the reconciliation process progressed, Stockton and Birmingham pushed the fact-finding component in new directions. Harvard professor of history Elizabeth Hinton and her students conducted archival research on Stockton’s history of police-community tensions. Though that work was still underway when this report was written, SPD’s procedural justice training team had begun incorporating material uncovered through the project into trust-building workshops for community members and officers. That material includes police reports containing hurtful and racist language used to refer to Stockton residents, and historical documentation showing how redlining practices contributed to residential segregation of racial and ethnic communities (see figure 3.2). The historical component of the trust-building workshops was designed to teach community members and police officers (from the command level to officers and professional staff) about the mistrust between law enforcement and communities. This material starkly illustrated the fraught intersection of race and policing throughout Stockton’s history. The conversation after the historical component involved small table discussions about how hearing that history made people feel and how law enforcement and communities could move forward. Those discussions also created spaces for communities to discuss events that had created mistrust. 42 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS FIGURE 3.2 Redlining in Stockton in the 1930s Note: This figure was used in Stockton trust-building workshops. Birmingham engaged in historical fact-finding by drawing upon the city’s civil rights history. In November 2018, the Birmingham PD, in partnership with the Birmingham Civil Rights Institute, the University of Alabama Department of History, and Sustainable Equity, LLC, expanded its cultural diversity training for new recruits to include a full day on Birmingham’s history and how it shapes police-community relations. This partnership drew upon advisory guidance from the National Holocaust Museum, the Montgomery (Alabama) Police Department, and the Nashville Public Library’s Civil Rights and a Civil Society program.58 The program included trust-building activities at the 16th Street Baptist Church, a detailed presentation on relevant local history by a University of Alabama history professor, a brief presentation by Chief Roper of BPD, and a period of discussion and reflection about the implications of the city’s history on BPD officers’ roles and their interactions with the community. Acknowledgement of Harm Each site’s chief of police formally launched the National Initiative reconciliation process by offering a public acknowledgment of harm, covering their departments’ histories of racial harm and ongoing NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 43 practices that harm and burden communities already dealing with high levels of crime and victimization. These initial acknowledgements occurred in meetings explicitly convened for that purpose. The first such meeting occurred in June 2016 in Minneapolis, where Chief Harteau had a series of meetings with stakeholders in the African American, Latinx, Native American, Somali, and LGBTQIA+ communities. Birmingham followed in August 2016, when Chief Roper held initial meetings with leaders of the civil rights movement, followed by members of the African American, LGBTQIA+, and youth communities. In October 2016, Chief Jones convened his initial meeting with Stockton’s Community Advisory Board CAB), where the Latinx, African American, faith-based, and youth communities were represented. Pittsburgh was the last site to hold a reconciliation launch event with an acknowledgement of harm, which it did in November 2018. Chief Harteau’s statement acknowledged that “hundreds of years of policing a racist status quo has left a legacy,” and she apologized for her department’s role in America’s “awful and racist past.” In Chief Roper’s statement to the veterans of the civil rights movement, he said: I often talk about how we police under the shadow of the civil rights struggle. In 1963, Birmingham was the most segregated city in America. The police department was the arms and legs of the most segregated city in America. We’d chase you all over the place, beat you up, throw you in jail left and right. We’re standing on the verge of history—we’ve done some things really well. But I know we have a long way to go. I’m committed, and my command team is committed, to being part of the solution and not part of the problem. Not only did we do things wrong in the sixties, we’ve done things wrong today. We’ve done things wrong today. It’s not a reflection on the brave men and women who do our job well to say that we have a long way to go. I’ve been wearing a uniform for 31 years now. When I look at what’s occurring in our nation and our city— the riots, the mistrust, the injustice—I know we can do better. Chief Roper’s statement included the promise to “do better,” and a key element of the National Initiative’s definition of the acknowledgment of harm is that police leadership enter the reconciliation process by publicly committing to changes such as policing more effectively and respectfully, reducing arrests and incarceration, focusing enforcement on key offenders rather than broad community-wide enforcement of low-level offenses, and implementing strategies that proactively repair and reduce harm. To support this, NNSC developed a matrix of “confidence-building measures,” changes to policy and practice the police chiefs could make and announce at the beginning of the reconciliation process to demonstrate their commitment to identifying and mitigating harmful practices. Such changes included felony diversion, warrant clearance, assistance expunging criminal records, and crisis intervention and de-escalation trainings. As an example of how these confidence-building measures could be announced, Chief Jones attended Stockton’s reconciliation kickoff session with the Community Advisory Board and announced plans to launch a warrant amnesty program and a youth diversion program. Changes discussed during the Minneapolis acknowledgement of harm meetings included an amendment to 44 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS MPD’s use-of-force policy to prioritize sanctity of life for officers and civilians, a requirement that officers intervene in and report incidents in which other officers use excessive force, and the ongoing procedural justice and implicit bias trainings. The acknowledgements of historical harm were ultimately delivered in conjunction with the listening sessions, described in the following section. Sustained Listening: Reconciliation Listening Sessions The core component of the reconciliation process was a series of listening sessions with the community. These sessions were intended to reflect a commitment to listening to communities and their diverse residents (focusing on those with tense and problematic relationships with the police), acknowledging past and present harms, and using new information to change policing practice in accordance with shared priorities and values. National Network for Safe Communities articulated the purpose of the listening sessions in a progress report to DOJ: “The goal of each focused listening session is to establish trust between law enforcement and communities to enable them to work together to consider policies, practices, and norms—both within law enforcement and within communities—to reduce violence and harm and improve public safety outcomes. The operational design of the listening sessions has been designed to create a cascade effect, with the slow, deliberate, and continuous incorporation of rank- and-file and community members as the work continues.” The listening sessions occurred on varying timelines across sites. This was partly because the reconciliation framework was new, requiring the NNSC team to be closely involved with its implementation in each site (this made it difficult to start the process simultaneously across sites). Police leadership in each site also required substantial prep work to ensure they grasped the purpose and requirements of the reconciliation process and were ready to implement it. Finally, unforeseen developments such as changes in police leadership and critical incidents delayed the reconciliation process in some cases and interrupted it in others. For example, Pittsburgh was ready to launch the listening sessions when Chief McLay resigned, and the transition in leadership to Chief Scott Schubert delayed that launch. Listening sessions began in Birmingham, Minneapolis, and Stockton between August and October 2016 (see table 3.1). Birmingham also initiated an important variant of the listening sessions when it launched its intensive small group circle process in October 2016 (see box 4). In April 2017, Gary became the fourth site to initiate listening sessions, and Fort Worth held its first session at the end of 2017. Pittsburgh was the last site to begin its sessions, which it did in November 2018. Because each NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 45 site implemented listening sessions at different intervals and held different amounts of sessions, their communities and police departments received varying degrees of exposure to them. Birmingham and Stockton had held the most listening sessions by the end of 2018, and Fort Worth and Pittsburgh the fewest. It is worth noting that the participating police departments had existing community engagement efforts before starting the reconciliation process, and in some cases, interview respondents seemed to confuse reconciliation listening sessions with community meetings in general. For instance, Chief Jones held “Courageous Conversations” in Stockton in 2016, which were not part of the National Initiative reconciliation process (unlike the reconciliation listening sessions, they were open to the public). Launching the reconciliation process required Chief Jones to reframe the Courageous Conversations initiative under the title “Listening in a New Way.” Moreover, though Pittsburgh was the last site to initiate reconciliation, a zone commander undertook a series of outreach efforts with youth after he had begun National Initiative activities, but before the reconciliation process. TABLE 3.1 Reconciliation Listening Sessions by Site Date of first listening session of listening sessions (through 12/2018) Areas of focus Site Birmingham 8/2016 10/2016 circles) 19 3-week intensive small group circles Internal PD, intimate partner violence (IPV) survivors, Latinx/immigrant communities; LGBTQIA+, neighborhood, youth Fort Worth 12/2017 3 Neighborhood Gary 4/2017 6 IPV survivors, neighborhood Minneapolis 8/2016 8 African Americans, clergy, Latinx, LGBTQIA+, Native Americans, neighborhood, youth Pittsburgh 11/2018 3 Youth Stockton 10/2016 20+ Community organizations, group violence intervention (GVI) client, LGBTQIA+, neighborhood, racial/ethnic communities, youth DEFINING LISTENING SESSION COMMUNITIES Communities involved in the listening sessions included people with intersecting identities, complicating their relationships and levels of trust with the police. Each National Initiative site developed a strategy for defining the community for each listening session. Sites did so in various ways, some of which are summarized in table 3.2. Many listening sessions were designed as means of engaging communities, identifying problems, and developing solutions to police relationships with the subpopulations discussed in box 2. Subpopulations included survivors of intimate partner violence 46 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS and/or sexual assault (Birmingham, Gary), the LGBTQIA+ community (Birmingham, Minneapolis, Stockton), youth (Birmingham, Minneapolis, Pittsburgh, Stockton), Native Americans (Minneapolis), and Latinx/immigrant communities (Birmingham, Minneapolis, Stockton). In addition to defining audiences by their subpopulations, sites used existing approaches and developed new ones to organize the participants for listening sessions. A common approach was to divide sessions according to neighborhoods where police-community tensions and concentrations of crime and violence were high. Police said in interviews that this approach created the opportunity to use the sessions as a platform for ongoing dialogue, given police departments are organized geographically. Other approaches included developing sessions by partnering with community organizations with credibility among residents (Stockton), holding listening sessions within the police department (Birmingham), and holding listening sessions with a group of GVI call-in attendees—those most at risk of violence—to acknowledge the harmful impacts of over- and underpolicing and hear about their experiences with the police department (Stockton; see box 3). How listening session communities were defined and whether the “right” people were included depended on who was responsible for scoping and running the sessions. The NNSC technical assistance team tended to play an active role planning and launching the listening sessions in each site, and sites that committed to holding many sessions had to develop processes that made doing so locally feasible. For example, the Birmingham PD had commanders set up listening sessions in their command areas, which Chief Roper prepared for by facilitating a mock listening session for the command staff. However, a change in police chiefs delayed this process. Birmingham ultimately used grant funding to hire a community activist and organizer as a community reconciliation liaison, who helped them engage with the community. Similarly, Stockton’s Community Advisory Board, which consisted of a diverse group of community members and advocates, played an active role setting up the listening sessions. BOX 3 Ceasefire Leadership Council Listening Sessions The Stockton Police Department has worked with the city’s Office of Violence Prevention (OVP) and the California Partnership for Safe Communities (CPSC) to involve Operation Ceasefire call-in attendees in listening sessions.59 These attendees are at high risk of involvement in gun violence (that is, at risk of shooting and being shot). Ceasefire call-in attendees were involved in listening sessions in two ways. First, Stockton has convened a group of call-in attendees committed to the project of keeping themselves alive and out of jail (and working hard to improve their lives) known collectively as the Leadership Council.” The Leadership Council meets regularly and has engaged in a series of listening NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 47 sessions designed to expose SPD and OVP to their narratives about the police. These stories, impressions, and feelings are documented, with the understanding that those narratives will be used to improve police training, policy, and practice as appropriate. Additionally, all Ceasefire call-ins are followed by three opportunities to participate in focus groups: one with Peacekeepers (outreach workers employed by OVP), a second with SPD that includes an acknowledgement of harm and an opportunity to hear participants’ perspectives toward and experiences with police, and a third that is a follow-up with Peacekeepers. These focus groups are also opportunities to identify potential new Leadership Council members. Several important themes have emerged from these listening sessions, particularly around people’s experiences with stops and searches in the streets, as well as parole/probation searches in their homes, which they noted can be particularly painful when family members are present. The listening sessions with the Leadership Council and Ceasefire call-in attendees reflect Stockton’s commitment to combining trust-building and violence reduction work, and to including those most at risk of committing and being victimized by gun violence in their reconciliation work. Source: Interviews with Stockton stakeholders. LISTENING SESSION CONTENT AND STRUCTURE The listening sessions took place in small groups (typically 20 participants or fewer) at community locations intended to help participants feel welcomed, comfortable, and safe. These included churches, community centers, and spaces in community organizations. Rooms were arranged for participants to sit in a circle. The police chiefs usually spoke for the police and delivered a version of the acknowledgement of harm at the outset of the session. They also often facilitated discussions, although nonpolice facilitators were used in some cases. In interviews, participants described these discussions as loosely structured and guided by the issues community participants raised. Everyone was afforded the opportunity to speak, though not every participant did. A Gary listening session participant said, One thing I like about it was the size of the session was small. That intimate atmosphere better promoted community, a person wasn’t rushed.” Participants described the chiefs’ role as active, but focused on listening rather than speaking, which participants generally appreciated. Each listening session was designed to include members of a particular community, although some community members reported attending more than one session. Conversely, police chiefs and leadership were present at many sessions. The exception was the Birmingham Public Safety and Equity Circle process, during which community and police participants engaged as a small group over a three- week intensive period, described in box 4. Participants of this process viewed it positively, and the group continued working together after the three weeks. However, implementing it was resource- 48 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS intensive, and the model had not been replicated in Birmingham or the other National Initiative sites when this report was written. Listening sessions surfaced a variety of themes and concrete issues. Table 3.2 summarizes some of the key issues that observers from NNSC documented in the initial listening sessions in Birmingham, Minneapolis, Stockton, Gary, and Fort Worth. The table offers a sense of the variety and nature of the issues that surfaced during the sessions. TABLE 3.2 Themes from Initial Site Listening Sessions Date and audience Themes Site Birmingham 8/16/16, LGBTQIA+ community hearing that a public official wants to hear and help the LGBTQIA+ community extremely unusual concerns about ways LGBTQIA+ youth engaged in sex work are policed o they fear the police and don’t feel protected by them o sexual assault cases involving LGBTQIA+ victims don’t go anywhere know your rights training for public PD training on LGBTQIA+ issues and misconceptions explore “hate crime” designation identify correct identities in public statements about crime and violence Birmingham 8/16/16, youth clarify the mission of the police department to young people, and make them aware of other resources that might exist clarify what policing looks like beyond crime reduction and arrests expand engagement by making sure white male officers reach out quantify interactions for police (metrics) Minneapolis 8/24/16, African American community need for frank conversations and to acknowledge history lack of explanation about why law enforcement is doing what it’s doing officers need to reach out, invest in building relationships to develop trust allowing bias to cloud how community is treated lack of de-escalation techniques used accountability for officers lack of empathy department culture must change; simply adding more officers of color is insufficient Minneapolis 9/7/16, youth police shootings and mistreatment during interactions hurt public trust o “fear of police is real” citizen stops that feel unnecessary lack of empathy and understanding of culture/social norms Minneapolis 9/14/16, LGBTQIA+ community officers need better understanding of what it means to be trans/LGBTQIA+; use of proper pronouns some officers don’t know how to handle a same-sex DV situation focus on youth; how school resource officers are chosen; are officers aware their presence is intimidating? NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 49 Date and audience Themes Minneapolis 10/20/16, Native American community lack of follow-up and slow response time to shootings union leadership accountability for officers/complaint process need to reach out to youth lack of officers who reflect the community o current practice alienates youth/gives them criminal records, which impedes making department reflect the community assign officers to Little Earth area and learn about the Native American community Stockton 2/22/17, Black Leadership Council deep-rooted distrust of police in the black community accountability for police shootings and dishonest officers need community policing, don’t know officers anymore want support for civilian oversight Stockton 3/1/17, high school students building respectful and positive relations with the community composition of department/hiring more African American officers role of police in immigration enforcement Gary 7/24/17, neighborhood need more compassionate officers need to communicate better what PD is doing/changing to community Fort Worth 12/8/17, neighborhood will make policy changes clearer online officers to match community oversurveillance differential treatment by race accountability for officers behaving poorly Source: NNSC notes and summaries from listening sessions. These themes include concrete problems (e.g., departments not reflecting community diversity, need for accountability mechanisms for officer misconduct), suggestions for changes (e.g., training officers to better understand the LGBTQIA+ community, communicate better with the community), and broader overarching issues (e.g., perceived disparities in police responsiveness to certain communities, need for more empathy from police). The variety of themes reflected in this subset of sessions shows that consistently documenting problems, prioritizing changes, and reporting back to the community can be a challenging process. Are you here in the community to keep us safe or are you here harassing us? That was a recurring theme for that night. Fort Worth listening session participant As the sites deployed various strategies for engaging communities, their approaches to the listening sessions evolved. In particular, the community-facing procedural justice and implicit bias trainings and 50 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS other community engagements often adopted the format of listening sessions in whole or in part. For example, in June 2018, Stockton piloted a new format of listening sessions called “trust-building workshops,” a fusion between a listening session and community PJ training. Community members sat in small groups with SPD officers, and the SPD captain facilitating them presented the procedural justice training history module as an entry point for inviting community members to share their experiences. Similarly, Birmingham Equally United engage d in reconciliation work in schools that incorporated elements of community implicit bias workshops and listening sessions. BOX 4 Birmingham Public Safety and Equity Circle Process A unique variant of the reconciliation sessions was fielded in Birmingham, where NNSC worked with Sustainable Equity to develop and pilot a sustained small group process. Eighteen participants (10 police officers and 8 community members) from the west and east side precincts and neighborhoods formed the group. Community members included people involved in Birmingham’s Black Lives Matter chapter, neighborhood associations, and churches. The group met three times a week from October 10 to October 28, 2016, and went on a weekend retreat in Montgomery, Alabama, where they visited the Equal Justice Initiative. The group had the following goals for those three weeks: Week one: remove barriers to trust by identifying who we (participants) are beyond labels. Week two: focus on the role of police, what officers and communities perceive police as doing now, and what they should be doing. Week three: identify tangible deliverables, next steps, and goals for operationalizing reconciliation moving forward. The emphasis was to come up with concrete next steps and measurable changes, beyond changes in “trust”/emotions alone. Sustainable Equity developed the pilot process based on the William Winter Institute for Racial Reconciliation’s Welcome Table community-building initiative. Initial listening sessions between the chiefs and community members helped identify specific areas of concern and improvement that the small group circles could address and solutions they could work toward. In facilitated meetings, guided by a written curriculum, rank-and-file officers partnered with community members to share personal experiences, build trust, identify common ground, and strategize on opportunities for local collaboration. During this three-week intensive process, the group developed a strategy for ongoing engagement and citywide expansion, although that expansion had not happened when this report was written. The circles allowed participants to overcome wariness and build mutual understanding. As one participant described, “You can tell that the citizens and police were kinda skeptical about what they NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 51 were willing to say and being receptive for what the other had to say. As each meeting kinds went on the more time we got to meet and sit and talk and open up about things it started to show everyone that it’s not about the uniform, the badge, the citizens. We are all human being we all have feelings we all have our own opinions, but we all want the same outcomes … as far as a better living … for all people in all communities.” A group of 10 to 12 participants from these circles continued meeting regularly after the small group reconciliation process, naming themselves Birmingham Equally United and undertaking several activities intended to advance the reconciliation work, such as bringing reconciliation programming to an alternative high school. The program involves a three-week trust-building workshop that incorporates material from the procedural justice and implicit bias trainings, as well as a version of the listening session structure. Sources: Participant interviews and internal National Initiative documentation of the equity circle process. Narrative Collection and Dissemination As indicated above, the listening sessions surfaced important narratives regarding communities’ views of the police and, to a lesser degree (given the role of police in these sessions was to listen), officers’ views of the community. The National Initiative reconciliation framework includes collecting these narratives and disseminating them to people who could not attend a listening session. Although themes had been synthesized to some degree when this report was written (most notably in Stockton), they tended to focus on areas for policy and practice change. When this report was written, none of the sites had undertaken communication activities to share the narratives with their communities. However, Chief Jones proposed that Stockton’s Community Advisory Board establish a Narratives Subcommittee to work with SPD to expand its collection and dissemination of narratives. This subcommittee’s goals would include the following: taking anonymized transcription notes at future listening sessions to highlight follow-up items, provide mutual accountability, and flag community and police narratives that should be captured work with SPD to develop a strategy for collecting officer narratives and exposing officers to community narratives (e.g., through video or audio) in ways that are emotionally evocative and informative work with SPD to develop a plan to ensure that narratives are collected with sensitivity to triggering communities and officer trauma 52 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Chapter 4: Changing Police Department Policy The National Initiative was based on the premise that improving public trust in police required new thinking and practice. Changing departments’ cultures required impacting their thinking and practices and institutionalizing new ways of policing in their policy architectures and departmental structures. Police stakeholders stressed that new policies were important for supporting the consistency and long-term sustainability of trust-building practices. This chapter focuses on how the National Initiative changed police department policy and practice across sites. It begins by describing how National Initiative activities generated ideas for policy changes. It then describes specific policy changes (as well as the processes developed to identify, prioritize, and carry out those changes) and concludes with changes to organizational structures that were consistent with the National Initiative’s goals. National Initiative Efforts That Inspired Policy Changes A goal of the National Initiative was to encourage police departments to change policy and practice to support the initiative’s three conceptual pillars and build and maintain community trust. Prompted by the efforts of National Initiative partners and participants (such as Yale Law School’s model procedural justice policies and the reconciliation process) and other National Initiative activities (such as the procedural justice and implicit bias trainings), departments did make changes to policy and practice. This section focuses on how National Initiative activities facilitated policy change. Procedural Justice Model Policies The Justice Collaboratory at Yale Law School developed a set of model procedural justice policies for police departments. By creating model policies, the National Initiative sought to inform policy changes in the six sites and provide a resource for police departments nationwide. As shown in table 4.1, the model policies were organized into the following four broad areas: transparency and public engagement procedural justice inside the police department officer interactions with communities practicing procedural justice in interactions with particular groups NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 53 The model policies were provided to participating departments in September 2017, and YLS staff visited each department shortly thereafter to brief police leadership on the principles and teach them to use them in a policy-change process. The principles of procedurally just policing were publicly released in April 2018 Quattlebaum, Meares, and Tyler 2018). TABLE 4.1 Principles of Procedurally Just Policing Transparency and public engagement PJ inside the department Officer interactions with their communities Practicing PJ with particular groups Procedurally just policymaking make policies publicly available subject policies likely to substantially impact community members to intensive public review solicit community input when making or revising policies communicate reason for policy decisions Use-of-force review: principles and policies draft specific and comprehensive use-of- force guidelines make policies and data publicly available employ sentinel event review and early warning systems Body-worn and vehicle- mounted cameras involve the community in body-worn and vehicle-mounted camera policymaking develop clear guidelines for body- worn camera activation develop clear guidelines for vehicle- mounted camera activation develop clear guidelines for film retention employ a presumption in favor of film release Model policy for workplace procedural justice promote an ethic of respect give officers due process and voice accommodate officers’ needs and preferences provide employees with support recognize employees’ contributions Model policy for performance assessment in procedurally just agencies encourage officers to act as guardians track and prioritize community trust reward procedural justice encourage a growth mindset make performance metrics transparent to the public model procedural justice at all levels of the department Model policies for police dispatch procedures provide supplemental training to 911 dispatchers incorporate the best available data and technology employ and expand alternative call systems and responses Investigatory and traffic stops limit investigatory stops to appropriate circumstances limit traffic stops to appropriate circumstances employ procedural justice in all stops De-escalation of conflict use de-escalation techniques in all encounters give officers de- escalation training publicly debate vehicle pursuit policy LGBTQIA+ individuals treat all people with respect ensure the safety of all people conduct outreach to LGBTQIA+ organizations Immigrant communities decouple federal immigration enforcement from local policing adopt inclusive and welcoming attitude toward immigrant members of the community Racial minorities developing training programs and policies that reduce the potential for racial biases to affect decisionmaking promote positive interactions between racial minorities and police officers Young people engage young people in the coproduction of public safety employ procedural justice in interactions with young people 54 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Policy Review and Analysis of Administrative Data In parallel with the development of the procedural justice model policies document, Center for Policing Equity staff reviewed the police departments’ existing policies to identify areas where policies could be strengthened or changed. Staff focused on the following five key objectives (based on CPE’s National Justice Database project): promoting unbiased policing improving community relationships reducing use of force strengthening oversight and accountability improving internal data collection and management The six police departments provided their policy materials by the end of 2015, which CPE supplemented with interviews with key departmental personnel. Policy reviews were provided to departments as completed on the following timeline: Birmingham: December 2016 Stockton: February 2017 Minneapolis and Pittsburgh: July 2017 Gary: January 2018 Fort Worth: July 2018 Table 4.2 summarizes the policy areas that CPE examined, categorized by the five key objectives. Across sites, policies on data analysis, self-assessment, and using that information for performance evaluation were uncommon. For example, five of the six departments had a written commitment to unbiased policing in their policies, but only two used this as a criterion for evaluating performance. Similarly, five of the six departments had policies related to pursuing activities to improve relations with specific communities, but only one had procedures for data collection and self-assessment in this area or used police-community relations as a criterion for evaluating performance. Every department had a written policy on use of force, but only one had a policy regarding the collection and analysis of use-of- force data. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 55 TABLE 4.2 Prevalence of Key Extant Policies Site prevalence Promoting unbiased policing Written commitment to unbiased policing Provided examples/scenarios of unbiased policing in policies Provided guidelines for practicing unbiased policing in policies Incorporated community input into unbiased policing policies Established procedures for reporting biased policing behavior Established procedures for data collection and self-assessment Used unbiased policing as performance evaluation criteria Improving community relations Written commitment to improving police-community relations Provided guidelines for improving police-community relations in policies Incorporated community input into police-community relations policies Pursued activities to engage specific communities Established procedures for data collection and self-assessment Used police-community relations as performance evaluation criteria Reducing use of force Written use-of-force policy Policy is restrictive (officers will always seek to de-escalate first) Provided clear guidelines on circumstances in which force may be justified Documented specific strategies to reduce use of force Implemented de-escalation trainings to reduce use-of-force incidents Established procedures for reporting use of force Data on use-of-force incidents collected and analyzed Strengthening oversight and accountability Early warning system (EWS) policy implemented Body-worn camera (BWC) policy implemented Guidelines on frequency and nature of reviews of EWS and BWC footage Established investigation and discipline process EWS and/or BWC used in performance evaluation criteria Improving data collection and management Established crime analysis unit or crime analyst embedded in department Clear, consistent procedures for collecting, analyzing data Used data to inform community-oriented policing efforts Used data to inform policy around use-of-force incidents Used data to inform policies related to racial disparities in enforcement Source: Urban Institute analysis of CPE department policy reviews provided to National Initiative sites. In addition to the policy review, all six departments agreed to participate in CPE’s National Justice Database, which CPE used to support changes to policy and practice by (1) analyzing administrative data on each department’s pedestrian stops, vehicle stops, and uses of force, (2) administering a 30- minute online survey to each department’s staff on department climate, and (3) surveying a subset of approximately 250 officers to determine how their behaviors related to their survey responses on department climate.60 These components were intended to provide different but complementary sources for identifying issues and areas for improvement. However, in practice, the policy review and 56 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS database activities proceeded on different timelines, and the results of the climate survey and administrative data analysis had not been released when this report was written. Training Feedback and Internal Procedural Justice During the procedural justice trainings, officers suggested changes to policy and practice that they felt were necessary. As noted in chapter 2, many officers had serious concerns about internal procedural justice, such as feeling like they did not have a voice or that rules did not apply to individuals in positions of authority. This feedback provided an additional impetus to examine ways of improving procedural justice through policy. A Minneapolis respondent described the values and practices that needed to guide internal procedural justice practices and policy by saying, “A lot of that has to do with consistency and process, transparency in processes and communication. So we’re doing a lot of work in those areas.” The internal procedural justice policies that Yale Law School included in its model policies document provided insights in this area as well. Reconciliation Listening Sessions As detailed in chapter 3, the reconciliation process identified areas of community concern that policy changes could address. The listening sessions were key venues for incorporating community input into the policy change process. The structure for police-community engagement involved treating changes to policy and practice as inputs (in the form of trust-building changes unveiled during initial acknowledgements of harm) and outputs. Listening sessions (the primary means of police-community engagement in the reconciliation process) generated ideas for changes to policy and practice. Some of these changes occurred quickly as community members raised issues that police leadership could address. For example, a Birmingham officer described hearing community views that led to a concrete change: “There was a perception from the LGBTQIA+ community] that we did not work as hard when one of them was victimized … From that session, we developed a policy, provided training on the policy, and we appointed a liaison to that community.” Similarly, Stockton PD began an effort to become a trauma-informed police department after engaging with survivors during the reconciliation process. One officer described how listening sessions prompted this effort: “I didn’t even know what it meant to be trauma-informed, really. I didn’t have crime survivors involved, even on my Community Advisory Board and really found out that a lot of folks NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 57 felt victimized twice by whatever the crime or the police contact and then the lack of being trauma - informed in the way we handle it. We’ve made some adjustments there.” Group Violence Intervention The NNSC helped the sites improve their responses to violence by implementing the Group Violence Intervention (also known as Operation Ceasefire). The GVI model uses the following strategies to reduce violence: engaging with those most active in gun violence (perpetrators and victims) leaning on those individuals to stop gun violence by communicating a credible moral message against violence during a “call-in” meeting or through individual engagement (“custom notifications”) giving those who receive the call-in message notice about the legal consequences of further violence delivering targeted enforcement consequences to those who continue to engage in gun violence offering services and assistance to those who want to desist from crime Several departments had implemented or had worked with NNSC to implement GVI before joining the National Initiative. Birmingham and Stockton continued implementing their GVI efforts (called the Violence Reduction Initiative and Ceasefire, respectively) while participating in the initiative. Birmingham held its first call-in meeting with active group/gang members on June 2, 2015, and added individual custom notifications later that year. Gary, Minneapolis, and Pittsburgh implemented their GVI efforts during the National Initiative period. Gary conducted its first call-in in April 2015, and Minneapolis in May 2017. Pittsburgh used custom notifications only (no call-ins) in its GVI effort and began them in April 2018. Fort Worth explored a related deterrence strategy, the Drug Market Intervention, but did not implement it. Birmingham discontinued the Violence Reduction Initiative after a change in city leadership and Chief Roper’s departure. Many site interview respondents noted that GVI and the National Initiative could be mutually reinforcing. Specifically, they felt that more effective antiviolence work would build trust by addressing the criticism that the police don’t care as much about harm done to communities of color or communities with high levels of poverty. They also thought that GVI’s focused nature (relative to 58 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS traditional approaches to addressing violence) was beneficial for building trust. However, some respondents in Birmingham and Gary noted that there was confusion about the distinction between the National Initiative and GVI, and many expressed a desire for more guidance on thinking about the connections between the two. (For more details, see NNSC [2018c], a case study on procedural justice, trust, and legitimacy in Group Violence Intervention activities in Gary.) Policy Changes This section describes changes the six police departments made during the project period through the end of 2018. It starts by summarizing the departments’ specific policy changes. It then discusses the processes the departments devised to review and change those policies (prioritizing areas for change and making those changes in a systematic and sustained manner). It concludes by discussing how the departments sought to institutionalize the gains they made during the initiative. Specific Policies Policy changes occurred throughout the implementation period, in part because the sites identified changes to policy and practice in various ways. Table 4.3 summarizes policy changes resulting from or influenced by National Initiative activities.61 Though the policy changes cover a wide variety of areas, there were some common themes. Many of the changes involved incorporating procedural justice language into relevant policies. For instance, Stockton incorporated procedural justice language into its departmental rules and regulations regarding conduct toward the public (rule 3.05) and fellow employees (rule 3.06). The language on conduct toward the public reads, “Members shall respect and protect the rights of the public and shall perform in a manner consistent with the principles outlined in the Department’s Mission Statement and shall ensure that they adhere to the tenets of Procedural Justice (Voice, Neutrality, Respect and Trustworthiness).”62 Similarly, the language regarding conduct toward other members of the department concludes, “Members shall not use language or engage in acts that demean, harass, or intimidate another person, and shall ensure that they adhere to the tenets of Procedural Justice (Voice, Neutrality, Respect and Trustworthiness).” Pittsburgh also added procedural justice concepts to its evaluation of field training given to officers and recruits and added procedural justice language to its policies on motor vehicle stops and warrantless searches and seizures. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 59 TABLE 4.3 Policy Changes during the National Initiative Implementation Period Policy changes Department Birmingham PD designated a sergeant as a liaison to the LGBTQIA+ community (September 2016) modified policy language to explicitly reinforce commitment to unbiased policing (2017) protection from abuse orders provided to all precincts by court rather than stored only in the precinct where the order was awarded, or rather than survivors being responsible for providing the order (2017) created new command-level position overseeing all community engagement (October 2018) Fort Worth PD revised general order on bias-free policing (February 2016) created standard operating procedures for Procedural Justice Unit (November 2016) began reporting use-of-force, arrest, stop, and discipline policies and statistics online (2017) revised general order on sexual assaults (July 2017) issued new order on racial profiling that reaffirmed the department’s commitment to unbiased policing (January 2018) revised use-of-force/force options and use-of-force reporting general orders (March 2018) revised departmental mission statement to add commitment to respect the sanctity of human life and preserve the rights and dignity of each individual in the community (March 2018) created new Police and Community Relationships general order, including role of Procedural Justice Unit (July 2018) added a duty to protect the safety and physical health of arrested and detained persons to the department’s Arrest Procedures General Order (August 2018) Gary PD None Minneapolis PD added transgender/gender nonconforming policy (June 2016) amended use-of-force policy to prioritize sanctity of life for both officers and civilians (July 2016) added policy requiring officers to intervene in incidents in which other officers use excessive force (July 2016) began tracking race and gender on traffic stops and other stops (September 2016) changed body-worn camera policy to require officers to turn on cameras as soon as they begin responding to 911 calls (July 2017) began reporting officer use of force, complaint, stop, crime, and arrest statistics online (2017) failure by an officer to comply with a lawful investigation of misconduct shall be deemed an act of misconduct (September 2018) Pittsburgh BP created ethics document (2016) added procedural justice concepts to evaluations of field training given to officers and recruits 2016) added order on transgender and gender nonconforming employees (August 2016) began posting policies online (January 2018) Stockton PD added procedural justice language to general order on how canines are deployed (May 2016) modified field training officer evaluations to add demonstration of PJ practices, and required supervisors to evaluate officer understanding of PJ in considering transfer/special assignment requests and promotional examinations (2016) mandated that officers receive annual mental health training mandated that officers make every attempt to mediate and defuse situations with people experiencing a mental health crisis (July 2016) added policy stating that “sworn personnel of the Stockton Police Department shall not stop, question, detain, arrest or place ‘an immigration hold’ on any person solely on the ground that he or she may be a deportable alien” (January 2017) added procedural justice language to rules and regulations regarding conduct toward the public and fellow police members (May 2017) tenets of procedural justice added to equestrian unit order (June 2017) created policy on release of body camera footage (July 2017) Unmanned Aircraft System policy created with community input (November 2017) 60 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Several departments changed policies to address forms of bias in policing. Fort Worth PD added an order on police and community relations to its general orders in July 2018 that reads, “In dealing with all people, each officer must attempt to make their contact one which inspires respect as an individual and professional, and one which generates the cooperation and approval of the public. While entitled to personal beliefs, an officer cannot allow individual feelings or prejudices to enter into public contacts. However, since an officer's prejudices may be subconsciously manifested, it is incumbent upon the officer to strive to avoid the influence of attitudes which might impair impartiality and effectiveness, keeping in mind that initial contacts leave lasting impressions both of the individual officer and of the Department as a whole.” Several sites changed policies to ensure officers engage more respectfully and appropriately with particular subpopulations, often those identified by the National Initiative. Formal Policy Review Processes Although National Initiative activities spurred policy changes on an ad hoc basis early on, the National Initiative technical assistance team worked with the sites to define and launch policy review processes to more systematically and strategically identify, prioritize, and execute policy changes. Specifically, NNSC staff engaged with the departments to identify the optimal policy review process through four areas of activity: (1) assessing existing infrastructure for reviewing policy and practice in each site, (2) using model policies, policy review recommendations, and any extant reports to determine site-specific priorities, (3) developing departmental infrastructure (such as policy review boards) for acting on these recommendations, and (4) where possible, connecting policy review boards with community counterparts and reconciliation listening sessions. A formal policy review process could allow for greater community input in policy development (as opposed to communities raising issues that departments translated into policy changes on their own) and a better feedback loop for communicating changes with communities and officers. Gary established an internal policy review board in late 2017 and was considering involving a citizen review board in its policy revision work. However, the policy review board stopped meeting when Chief McKinley left the department. As the chief transition progressed, GPD reduced its policy review group from 15 members to 4 to streamline the process at NNSC’s suggestion. The GPD’s policy infrastructure needed improvement: it had adopted policies from other departments that left it with numerous inconsistent and contradictory policies, and there was uncertainty regarding which versions of some policies were the most current. National Network for Safe Communities therefore recommended that Gary focus on defining its policy process by establishing bylaws for the policy review board, guidance for community input, and guidance for PD input; creating a directives development NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 61 policy; removing duplicate policies and reconciling contradictory policies; and posting policies online and making sure they could not be edited or otherwise altered. Birmingham PD convened a “council of captains” to undertake a policy review. The council was divided into five teams that were each tasked with reviewing a subset of Yale’s model policy principles. The teams provided an internal report to Chief Roper in July 2017 regarding the areas they believed extant policies adequately addressed, which should be addressed in policy changes, and how the model policies might be adapted for Birmingham. However, this process stopped when Chief Roper stepped down in early 2018, and a new process had yet to be established by the end of 2018. In late summer 2018, MPD convened a four-person policy team tasked with assessing the model procedural justice policies, the CPE policy review recommendations, and the recommendations for policy improvements the Office of Community-Oriented Policing Services provided the MPD. The policy review team incorporated community feedback from listening sessions and worked to involve the local ACLU chapter in their process. In Pittsburgh, members of the Procedural Justice Committee (a successor to the procedural justice training team) worked with NNSC to identify a list of policy areas to focus on revising in 2018. However, turnover among officers designated to serve in the policy unit delayed progress, and the hope was that assignments made in late 2018 would accelerate this process. In some sites, interview respondents said that departments needed more capacity to move policy reviews forward at the scale the National Initiative called for. The PBP reviews its policies annually, but one respondent noted that their policy unit consisted of a single person who was reviewing the policies largely to ensure they complied with accreditation requirements. Fort Worth made several important policy changes and established structures for incorporating more community input by launching the Chief’s Advisory Board and Policy Review Committee in 2017. These groups were instrumental in helping the department respond to the negative public reaction to a video of FWPD officers arresting Jacqueline Craig after responding to a confrontation between her 8-year-old son and her neighbor. Although stakeholders indicated that the group became inactive for a period in 2018, Chief Fitzgerald made plans to have the CAB meet bimonthly (rather than quarterly) in 2019. Stockton developed a robust policy change process with a leading role for community members. Stockton’s Policy Review Subcommittee, a subcommittee of its Community Advisory Board, began formally meeting in March 2018. The themes found in the reconciliation listening sessions guided its work, although it was not required to restrict its recommendations to those themes. The Stockton policy review process built on work that began in 2017, when SPD engaged the Community Advisory Board to develop policies for deploying drones.63 Though the board predated the 62 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS National Initiative, it assumed a much greater role in the policy review process because of Stockton’s participation in the initiative. An overview document prepared by SPD explained the subcommittee’s purpose: The Stockton Police Department is committed to ensuring that our policies are in alignment with what will best serve the community, with special attention to the communities that trust us least. Part of that process is for us to involve the community in helping craft and change policies and practices. We’ve worked to condense the many pages of notes into some broader themes which we would like to guide your consideration of possible changes to our policies and practices. Any changes will require a conversation about what’s plausible based on our resources, statutory requirements, and our needs. We hope that these conversations will help model the kind of respectful engagement that allows for us to combine law enforcement expertise with commun ity experiences and desires. The Policy Review Subcommittee begins the policy review process by developing recommendations for policy changes that it submits to Chief Jones for preliminary review. Those recommendations go to an internal SPD policy review committee (which includes a legal team) that passes policy changes back to the chief for final review. A Stockton police respondent noted that officers are hesitant to involve citizens in policy reviews, but that they became comfortable with CAB members through the listening session and procedural justice trainings. The listening session themes provided to the Policy Review Subcommittee included the following: Promoting transparency and communication. There was a sense that the department was not forthcoming with information regarding follow-up on criminal investigations, disciplinary investigation processes and outcomes, collected data, and departmental priorities and actions. Addressing trauma. Participants noted a lack of trauma-informed outreach by SPD to victims, witnesses, and families of those accused of serious (and particularly violent) crime. Discomfort with traffic stops. Participants considered traffic stops among the most difficult and stressful interactions between the police and the community. Communities and police are made uneasy by these encounters, and community members often do not know their rights and responsibilities. Discomfort during searches. There was a sentiment that home and vehicle searches are sometimes done vigorously and that officers damage property and create messes. Guided in part by this information from the listening sessions and by SPD’s comparison of its general orders with Yale Law School’s model procedural justice policies, the Policy Review Subcommittee selected policies related to mental health, parole, and probation searches and NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 63 investigations of deaths as the first three policy areas to review. By mid-2018, it had provided the results of its review to the chief, and in the second half of 2018, it reviewed policies in the areas of use of force, arrest procedures, transfers, arrests of juveniles, mobile field force, expressive activity, and the field training program. Institutionalizing the National Initiative Work within Police Departments The six police departments recognized that sustaining and expanding their trust-building work would require making institutional changes as well as policy changes. To that end, the departments made several changes to their departmental structures. Fort Worth PD and Minneapolis PD recognized the value of the procedural justice training teams for building trust and sought to formalize them as procedural justice units. Minneapolis PD created their procedural justice unit in response to the site coordinator’s and procedural justice training team’s early concerns about sustaining the procedural justice and National Initiative work. Chief Harteau approved a four-phase plan for embedding procedural justice in MPD in December 2016 and established the Procedural Justice Unit under the larger Community Outreach and Collaboration Advancement division. This division worked to integrate procedural justice into department activities by establishing “PJ ambassadors” at each precinct, who are tasked with implementing procedural justice concepts into protocols and leading monthly roll call trainings on procedural justice topics. In 2018, MPD decided to require that each precinct develop a procedural justice plan. These plans are intended to incorporate procedural justice values throughout all department trainings and operations, from internal review and accountability processes to onboarding for new recruits. An MPD overview of the process describes its purpose: “Developing PJ Precinct Plans does not mean starting from scratch; rather, it allows us to put a name and an emphasis on the great work that is already happening while encouraging new applications of these ideas. As we build this community of practice, we’ll see more and more applications of voice, neutrality, respect, and trustworthiness in the pursuit of these goals.” The precinct plan process emphasizes changes to practice, recognizing that many desired changes don’t require modifying policy. Minneapolis also developed a community navigator program, hiring six civilian community navigators to work with the African American, East African, Latinx, Native American, LGBTQIA+, and intimate partner violence survivor communities, respectively. Fort Worth PD was not as successful developing a procedural justice unit. The unit’s responsibilities were codified in FWPD policy: “The mission of the Procedural Justice Unit is to provide training and support to the Fort Worth Police Department and citizenry that enhances internal and external police 64 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS legitimacy thereby increasing trust, reducing crime, improving officer and public safety and help achieve the goals of our department.” However, FWPD’s procedural justice unit faced staffing challenges: it was reduced from five full-time trainers and one part-time supervisor (capable of training when necessary) to two full-time trainers and one part-time supervisor. As a result, Fort Worth respondents described the unit as largely defunct in the final survey. In Pittsburgh, the PBP and the Department of Public Safety created a Multicultural Unit, conceived as an effort to build trust between immigrant communities and various arms of city government, including the police. The Multicultural Unit’s efforts focus on three goals: communication and language access, outreach and education, and multicultural trainings. Its efforts also involve establishing a Public Safety Academy to be taught in Spanish, with input from the PBP procedural justice trainers (sometimes referred to as a procedural justice unit, though this designation was less formalized than in Minneapolis and Fort Worth); advising in the training of “Community Navigators” on immigration law and rights; providing educational opportunities in a variety of foreign languages on policing and public safety in Pittsburgh; and multicultural trainings for PBP to support officer safety and efficacy while interacting with diverse communities. Pittsburgh also designated 18 neighborhood officers to work with specific communities. (For more details, NNSC [2018a], a case study on Pittsburgh’s adapted procedural justice training and outreach to immigrant communities.) The PBP also assigned its lead PJ trainer to a permanent role at its academy so that he could incorporate procedural justice themes into existing departmental training. By the end of 2017, these efforts included adapting PBP’s patrol doctrine and use-of-force/de-escalation training to incorporate procedural justice. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 65 Chapter 5: Implementation Findings The National Initiative work was generative and iterative, and innovation occurred in key areas such as community trainings, internal procedural justice trainings, the reconciliation processes, and policy changes. The initiative was an experiment in improving police-community relationships using a variety of approaches, methods, and messengers. It presumed that police-community trust could be improved by first drawing on knowledge about research and practice and then providing training, technical assistance, and guidance to departments on translating this knowledge into practice. This process was not seamless, easy, or consistent across sites. The implementation evaluation yielded several key findings regarding relationships between the National Initiative partners, participating departments, and communities. Those findings included the following: The National Initiative partners had to overcome skepticism that outsider experts were sufficiently informed on local contexts and could provide insights that would bear on police- community relationships. Agreeing about the presence and scope of the problem was a challenge in some departments. Implementing the National Initiative required capable site coordinators and delegating the work to core teams, such as the procedural justice training team. Establishing consistent communication between the National Initiative team, police departments, and communities was difficult. Turnover within the National Initiative team in the initiative’s early stages exacerbated these difficulties. Creating a peer community among the six sites assisted implementation. Sites benefitted from consulting and conferring with peers in other sites as well as in nonparticipating cities, such as the Chicago PD academy trainers, who helped implement training. Peer exchanges improved site partner morale and consistency and spread innovation. Procedural justice was the orienting concept for National Initiative work within the police departments: all three trainings were named with the PJ shorthand (including the implicit bias component), and organizational structures created to advance the work used names such as the PJ unit.” Communities became involved in the National Initiative more slowly than police departments, in part because community-focused components like the reconciliation process and the 66 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS community-facing trainings were implemented at later stages, but also because communication with the community was not as strong as police and community stakeholders wanted it to be. Because training was one of the initiative’s first activities and was so resource-intensive, it was primarily focused on the police departments in the initiative’s early stages. Only later, as the community trainings and reconciliation process began, did partners and departments engage more robustly with communities. The National Initiative partners had to navigate police departments’ skepticism toward academics,” which several police respondents used as a shorthand for the partners. One officer said, Cops do not trust academics. These are people coming and telling me how to do my job without having done it.” Managing officers’ perceptions of outside experts was a challenge to translating those experts’ recommendations into policy. One police respondent in an early survey described their perceptions of policy recommendations: “From my perspective it’s kind of felt like … a whole bunch of people who sit around and think about what you do every day are going to give you advice about how to improve all this stuff. And we’ll have people who may or may not really understand what their suggestions mean, come down and give us suggestions.” One of the National Initiative’s key ideas was particularly difficult to communicate to command staffs: that low levels of trust in the police in communities of color and communities experiencing high levels of crime and violence resulted from harmful, disrespectful, and alienating policing practices. In some instances, police audiences received this information poorly. As a Fort Worth respondent said, The] message was perceived as ‘you guys are all doing it wrong and you’ve been doing illegal things. Your traffic stops are not legitimate.’ Regardless of how the message was intended, that’s not how you get trust. It turned a lot of people to ‘it’s a bunch of academics trying to tell us what to do.’” Multiple respondents in Fort Worth and Stockton described reacting negatively to the assertion that many police stops were unlawful. Similar challenges emerged from the MPD’s rank-and-file when National Initiative work began, which coincided with citywide protests in response to an MPD officer’s killing of Jamar Clark (see box 6). The first National Initiative project director, Dr. Tracie Keesee, had a credibility that was critically important in building relationships and establishing trust with the police departments. She departed roughly a year after implementation began, and many interviewees mentioned that this, as well as the substantial turnover in the National Initiative team more broadly, created communication issues between the National Initiative and the police departments. Turnover was also an issue within the six police departments. Chiefs changed in four of the six sites between the beginning of National Initiative NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 67 work and the end of 2018, which delayed and disrupted National Initiative activities. Fort Worth’s National Initiative coordinator also changed twice, and speculation over whether Chief Fitzgerald would leave in early 2019 created additional roadblocks toward the end of the grant. Though communication problems never disappeared entirely in the interviews, police respondents in later waves of interviews were much more likely to describe the working relationship and communication with the National Initiative as positive, supportive, and effective, suggesting that the relationships generally improved over time. As a Gary police respondent said of the National Initiative team several years into the work, “They’re more like family at this point.” Site coordinators offered a key form of leadership, as they were responsible for liaising with the partner organizations operating the National Initiative and all the local partners within and outside the police department, as well as advancing implementation of all the National Initiative components. This was a tremendous amount of responsibility for one person, especially considering that many of the site coordinators had other responsibilities, as in Gary, whose site coordinator also coordinated the Gary for Life antiviolence initiative. Site progress was generally greater where the work was shared. The procedural justice training teams were effective in this regard, and not merely in carrying out the labor- intensive work of training all the officers in the three-part training continuum. The procedural justice trainers also became experts in procedural justice subject matter, and in many sites, team members became leaders in reconciliation activities and policy development. The National Initiative work became important to many of the trainers on a personal level, which motivated them to champion the work within their departments. On the community side, community interview respondents in the baseline interview wave described being receptive to the National Initiative. However, several issues arose early in the implementation period regarding communication with the community. Interview respondents frequently said that communication with the community regarding what the National Initiative was doing was not strong, and that they did not know what was happening. This sentiment was more common in the first two interview waves than the third, but it still arose in the third wave. Some community members expressed frustration at not being informed of important developments such as Keesee’s departure as National Initiative director. Some community respondents perceived this lack of communication as a lack of transparency. As a Pittsburgh community member noted , “Transparency builds relationships, and when you can be transparent and proactive in your transparency, people feel like you see them as being important enough to know that. ‘Oh, only important people get that information’—that’s how it looks now.” 68 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS Implementing the procedural justice training for officers was resource-intensive for the police departments, meaning early training accomplishments were achieved within the departments and community members might not know much about them. Community respondents in several sites expressed that the National Initiative was more focused on internal work with the police department than on the community. One Pittsburgh respondent expressed this feeling by saying, “I think part of this initiative should be about working with the police and community together, not just the police department, so as we move forward, not just working with police by themselves and the community by themselves but bringing them together.” Of course, the reconciliation component in particular was designed to do exactly this, but reconciliation activities in the sites didn’t start intensively and publicly until at least a year into the initiative. Most people want to live in a city where if a cop pulls them over, they don’t have to take a deep breath. Gary resident BOX 5 Police Shootings in Minneapolis Lethal encounters between police and community members are the most painful flashpoints in police- community relations. The National Initiative sites were not immune to this, and no site had a more painful experience of this kind during the initiative than Minneapolis. The shooting of 24-year-old African American Minneapolis resident Jamar Clark in November 2015 launched widespread protests,64 culminating in an 18-day occupation of the 4th precinct police station.65 The community criticized both the police and city’s immediate response to the protests, decrying MPD’s use of chemical irritants, marking rounds, and riot gear,66 as well as their failure to protect protestors when five protestors were shot immediately outside the 4th precinct station by three masked men who appeared to be white supremacists.67 In March 2016, Hennepin County Attorney Mike Freeman announced that no charges would be filed against the officers,68 leading to further peaceful protests, and federal prosecutors have since declined to pursue civil rights charges.69 The shooting of Philando Castile in June 2016 has elevated scrutiny of MPD, even though Castile was shot by officers from the police department of Falcon Heights, a Saint Paul suburb. In July 2017, an MPD officer fatally shot Justine Damond, which led to Chief Harteau’s resignation. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 69 Minneapolis PJ trainers described the challenge of administering conceptual procedural justice soon after the Jamar Clark shooting. A Minneapolis PJ trainer described the work they had to do to ensure that officers were receptive to the training in the aftermath of the shooting: “We just cleared that up immediately, saying ‘this is not a response to the Jamar Clark shooting, we are one of six cities that was put in for the pilot.’ I just put it out there right away, so people would know this was proactive and not reactive, and they reacted well to that because you can’t have people thinking this is a punishment.” In fact, the trainer went on to say that delivering CPJ so soon after this incident may have been ideal timing, as the trainings gave officers a space to “vent.” Findings on Training Implementation Stakeholders across all six National Initiative sites considered the implementation of the procedural justice and implicit bias trainings to be a major success. Key findings from the assessment of training implementation included the following: Ensuring the trainers were “credible messengers” was a key to success. This was necessary to overcome officers’ resistance to such charged material as the racist history of policing, understanding the perspectives of community members who mistrust the police, and the operation of biases in one’s thought processes. Delivering 24 hours of training to every sworn officer in a police department placed a heavy load on the trainers and the departments. This strain was greater for trainers who retained other job responsibilities concurrent with training delivery. Extending the trainings to the community was an important innovation, but placing uncompensated responsibilities on community volunteer trainers was a barrier to doing so. Issues with internal procedural justice could impede officer uptake of the procedural justice concepts in their interactions with community members. Officers were initially skeptical of the training, reflecting what respondents described as a “wait and see attitude” toward the National Initiative and the feeling that it was a punishment or the first step in a federal takeover or intervention in the department. The training was the first in-depth exposure officers had to the National Initiative and its intentions, and as such, it was a key intervention for explaining its principles and developing buy-in. Intentionally selecting “credible messengers” to deliver the training was critical to overcoming this skepticism and making the trainings successful. As one interview respondent said, “The suggestions 70 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS from the National Initiative of picking the right people for the training is key. That is very important. I don’t know how many evaluations we had where people said, ‘You really picked good people to do this training,’ that either because of our work reputation or just how we’re able to relay the information and present it in such a manner that people really appreciated that.” A Fort Worth trainer described the problems that arose when a trainer did not have this credibility by saying, “The people teaching it have to be credible. The legitimacy flows internally and if the people trying to teach it aren’t seen as legitimate nobody wants to hear what they have to say.” The trainers themselves noted that peer exchanges with other sites and support from the CPD trainers were key in preparing them to successfully deliver the trainings. Interview respondents said delivering 24 hours of training to all officers over two or more years was heavy lifting for the training teams and that serving as a member of the training team was demanding. Members of training teams who were only able work as instructors part time, as was initially the case in Fort Worth, were most challenged to manage the demands on their time. The heavy demands on trainers resulted in turnover in the training teams. Only one of the initial Fort Worth instructors was on the PJ training team when TPJ concluded, and only one trainer from each of the initial Gary and Pittsburgh CPJ trainer cohorts remained when implicit bias began. Some of this turnover owed to promotions (two of the original Pittsburgh trainers became zone commanders). Several training team members continued serving as trainers after being promoted and balancing their new responsibilities with their responsibilities as trainers could also be challenging. All but one of the Birmingham trainers for the combined CPJ and TPJ trainings were patrol officers, whereas all but one of the IB trainers were in the Detective Bureau, which made finding time to teach the classes more stressful. Respondents described interacting with officers who indicated that procedural justice language and conduct was beginning to permeate officer thinking and behavior after trainings. A Gary trainer reported, “One of the things that happens with our agency when we instruct PJ we'll have officers come back—they'll kind of tease us if they see us and go ‘hey, watch me PJ this guy.’ They're giving that guy the voice. They'd given him his respect. They're being neutral … So, our guys are doing it even when they're just trying to be smart about it.” One concept from the CPJ training that interview respondents suggested had taken hold was the idea of a “community trust bank,” in which every interaction between an officer and a community member makes either a deposit to or a withdrawal from an overall community trust bank, and that a low or negative balance in the trust bank can give rise to community protests and increases in crime rates. A Pittsburgh trainer described an illustrative interaction by NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 71 saying, “People are coming up to me saying, ‘hey I just put a deposit in the bank account,’ and I’m like what? And they’re like ‘the community bank account commander, you know that thing you taught us?’” Procedural Justice trainers from multiple cities said that officers commonly felt that the procedural justice concepts introduced during CPJ were common sense and simply defined good policing. However, many of the trainers who described this reaction said that the procedural justice framework was nonetheless valuable because it put a name to this type of good policing, thereby elevating it. Officers in several sites considered CPJ’s being a refresher or reaffirmation of good policing as its most valuable element. A Birmingham PJ trainer articulated this common sentiment by saying, “A lot of officers feel that they’re doing PJ work already. But, they find it affirming rather than frustrating.” Still, the open text responses to the training surveys indicated that some officers took the opposite view, perceiving the common-sense nature of the procedural justice concepts to be a waste of their time, or even insulting. The trainers in the National Initiative departments were very impressed with the quality of the implicit bias curriculum and felt that its quality was critical to conveying the material successfully. As a Fort Worth trainer said, “The biggest strength is that this is helping everyone involved; community, supervisors, officers in the field to recognize: am I seeing this for what it is or what I think it is? When we develop that emotional intelligence to manage our own perceptions, it helps us create empathy. I see this as a wonderful tool to develop emotional intelligence. It helps us manage our own perceptions. When we can do that, we can do that with other people.” As with the procedural justice concepts presented in the first two trainings, trainers believed that the concepts stuck with officers after the training. A Minneapolis trainer noted, “Even the ones who left saying this is stupid, I still hear them joking about it. ‘Hey, don’t say that, you’re being implicit biased.’ And even if they’re joking it’s still in their head.” Departmental handling of internal procedural justice issues was seen as inextricabl e from the use of procedural justice in officer interactions with the public. As a Minneapolis respondent described, “Cops are struggling internally because they aren’t being treated well in the department: discipline, transfers, selective discipline of one officer versus others leads to a perceived unfair approach to officers depending on who they are.” A Fort Worth police respondent summarized the importance of addressing internal procedural justice to the overall success of the National Initiative by saying, “In order to expect officers to model behaviors in PJ classes, it has to be modeled from above. If it’s not modeled from above, the officers will recognize that and it’s kind of a lost cause.” These views are consistent with emergent research from the United Kingdom on the importance of internal procedural justice (referred 72 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS to there as “organizational justice”) changing police practice and culture (Aston, Murray, and O’Neill 2019; Bradford et al. 2013; Bradford and Quinton 2014). Trainers in four sites said participants considered the historical component one of the most valuable parts of the training. A Birmingham PD respondent discussed the importance of exposing all officers to that history early in their careers, saying, “I’ve always known that we have had a very bad history with race, stemming from the civil rights era. I didn’t necessarily have a good idea of the history, really, until I got to a position of authority. That history isn’t taught to new recruits, but I think it’s important because, as a patrol officer, I dealt with people who had been mistreated by the police, and I should have known about that as a rookie.” Covering this history in the procedural justice trainings had the added benefit of exposing all the officers in the participating departments to the facts of the historical harms that were addressed in the acknowledgments of harm and other reconciliation activities. Officers responded to the training by expressing a desire for reciprocity from the public. A Fort Worth trainer described the nature of this request, saying, “One of the blowbacks of PJ training is that officers are saying why are we getting trained. Why do we have to change, and the community doesn’t?” A Gary trainer described a similar dynamic, saying, “Their main concern is what is the public going to do to make changes as well. They would say, ‘we’re taking this class, and this is bettering us and making us better police officers. I have no problem taking the class, but what is the public doing on their side to make a change as well?’” The desire for the community to be exposed to the training came up in many training survey responses, which were split between suggestions that the community also receive the trainings and suggestions that the community receive the trainings instead. (Some officers and trainers also suggested that community members participate in the officer trainings in some way.) Community coleadership of the community trainings (which emerged in response to this officer feedback) was important. However, both police and community participants noted issues with the effort being asked from unpaid community volunteers. As a Birmingham community member said, “I think it's such a great opportunity, I just think a lot of times when these kinds of initiatives come down that the community is needed to participate in them for them to be successful but a lot of the time the organizers and community members that are brought in to help too often are taken for granted. Nobody thinks about their time and their energy being placed in it and the need for them to maintain their livelihood … I know as an organizer people want me to do beaucoup things for free.” In Birmingham, many of the community volunteers who initially engaged did not continue for this reason. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 73 Findings on Reconciliation, Police Department Culture Change, and Sustainability Reconciliation processes were intended to build and repair police-community relationships, facilitate departmental policy changes, and ensure that changes in the six sites were meaningful and long-lasting. Key findings from the implementation assessment regarding reconciliation, policy change, and sustaining the National Initiative’s work in the sites included the following: Successful reconciliation listening sessions required that police be open to hearing community perspectives and that they not react defensively to angry or critical statements. Leadership support, most notably from the chiefs, was necessary to model the National Initiative. Such support is critical for modelling procedural justice to officers, speaking for the department in reconciliation activities, making key policy changes, and providing the organizational support necessary to advance the National Initiative’s work. Changes in chiefs could interrupt progress. Turnover in police chiefs is common, and four of the six sites experienced a chief change during the implementation period. Such changes were particularly disruptive of the reconciliation and policy change work, delaying progress even when the new chief was supportive. Deliberate selection of training teams created internal champions for the National Initiative concepts. The trainers became designated procedural justice units in some departments, and training team members became involved in other National Initiative activities such as community trainings, department policy reviews, and reconciliation listening sessions. Support of police leadership was described as necessary for the National Initiative to succeed, and for sustaining that success. Based on stakeholder interviews, there was strong support from leadership in every site except Fort Worth, where the lack of leadership support was described as a significant problem. A common theme in the interviews with community members was that views of the police were somewhat distinct from those of the chief, which were often more positive. The apparent strengths of the relationships between the chiefs and the community were an asset to the National Initiative, but also a potential source of instability. Four of the six sites experienced a change in chiefs between joining the initiative in 2015 and the end of 2018. Interview respondents emphasized that leadership support needed to include dedicating the necessary resources to National Initiative activities, but also modelling the concepts to change their departments’ cultures. Multiple respondents considered the frontline supervisors’ modelling to be critical, describing it as a primary vehicle for the 74 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS essential work of integrating procedural justice and National Initiative concepts into the departments’ organizational cultures. As a Pittsburgh trainer put it, “If you have supervisors doing things that are not PJ and quite frankly undercutting what PJ is, that spreads like a virus.” In most of the reconciliation listening sessions, the chief of police delivered the acknowledgement of harm and facilitated the discussions, and many reconciliation participants emphasized the importance of the chief’s presence. A Gary police participant reflected on how important the chief ’s presence was, saying, “People really wanted to hear from the chief. I don’t think they would accept answers from a sergeant in patrol or even [a member of the command staff].” The centrality of chiefs to the reconciliation process involved some risk: turnover in police leadership delayed or disrupted progress in implementing the reconciliation process in several sites. Many police and community respondents who participated in the listening sessions were impressed by them and felt that they went well, emphasizing that participants were mutually respectful. This allowed participants to exchange the views necessary for finding common ground. However, this is not to say that respondents felt the sessions were easy. Many community participants described coming to the listening session with some degree of wariness, sometimes feeling skeptical of police participants’ sincerity, and sometimes anticipating an unproductive airing of hostility toward the police. Several participants said the listening sessions began with anger from the community, and claimed the police had to be ready to hear raw or difficult things. They stressed the importance of the police being able to listen to these things and respond without defensiveness, which allowed the sessions to proceed to a positive place. A Stockton community participant described how police participants reacted to criticism, saying, “They were listening. They didn’t try to minimize any of our stances. They were attentive. They were listening.” Setting up listening sessions with community members who had very negative views of the police required attentiveness to creating safe spaces for them. Community respondents noted that engaging youth in the listening sessions was particularly difficult in this regard, because youth often had the most negative recent experiences with law enforcement and were the wariest of sharing a space with officers. For instance, interview respondents described difficulties making LGBTQIA+ youth in Birmingham feel comfortable having police in their space, although their listening session was ultimately successful. An issue that arose in multiple sites was that officers attending listening sessions in full uniform alienated youth participants in particular. This was also an issue at the first Birmingham listening session, where half the participating officers came in uniform despite having been asked not to do so. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 75 Listening session participants who Urban interviewed emphasized the importance of the acknowledgement of harm to the process, but also the challenges of getting it right. A Stockton officer reflected on this, saying, “Also surprising to me has been how when the acknowledgments of past harms are made, how that just opens up these floodgates of communication that we have been struggling with for years and years. Decades, really. Where we kept meeting with the same people or the groups, and we did either town halls, public forums, separate meetings, private meetings, and we weren’t really making a whole lot of significant progress and trust-building.” A challenge of the acknowledgement of harm was ensuring that it included current practices in addition to historical ones. One Minneapolis community participant was frustrated by the perceived limitations of the acknowledgment, saying, “We know it’s derived from slave patrol. Acknowledge the things they’ve done to us. I’ve had police do terrible things to me.” Several police respondents expressed that the listening sessions were creating connections with people in the community they hadn’t previously had connections with. As a Gary police respondent reflected, “It was a dialogue opened with parts of the community we don’t usually hear from. We have monthly meetings in some communities that are just not well attended. None of the people I saw at the sessions were people I usually see at meetings.” Stockton police respondents found that their listening sessions brought to their attention community groups who should be engaged, including several influential groups that hadn’t been on their radar. I have issues with the word ‘reconciliation,’ because … we’ve never had the relationships we could or should; we’re trying to get to some place we’ve never been. And I open meetings with that. I would challenge people to tell me a time when relationships between police and community have ever been good. We’re not trying to get back there; we weren’t there. Minneapolis police respondent Community members had more concerns than the police about whether the right people were included in the reconciliation sessions. For example, some community respondents were concerned that people who were critical of the police or would challenge them were not included (this was a particularly pointed critique of the Fort Worth listening sessions), or that attendees were not representative of the community. However, not all respondents saw this as a concern: one respondent 76 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS recalled that the National Initiative didn’t involve the police department in organizing the first reconciliation meetings to ensure that they were not attended solely by people with positive views of the police. Others thought the sessions did not reach the community extensively enough, and that the community should have been involved in planning the process before it launched, the type of concern that Birmingham PD sought to address by hiring a temporary community reconciliation liaison. A few community members also said that for listening sessions involving sensitive communities such as youth and survivors of sexual assault, departments should invite members of those communities themselves in addition to advocates and service providers. Interviewees expressed that developing listening sessions and other reconciliation activities for specific subpopulations (as defined by DOJ) had benefits including the National Initiative partners’ development of specific engagement guidelines, and higher engagement with communities the police might have had little previous intentional engagement with (this was particularly the case for the LGBTQIA+ community and survivors of intimate partner violence in several sites). As a Birmingham police stakeholder described, “In Birmingham, our historical context has always been black-white. But we need to think beyond that, because we have a diverse, multicultural community.” However, stakeholders also noted some shortcomings of the National Initiative’s subpopulation approach. One concerned the National Initiative’s and DOJ’s definition of subpopulations, and a perceived presupposition about what these communities’ concerns were. Some stakeholders considered the consolidation of Latinx/immigrant communities into a single subpopulation problematic, especially in cities with large Latinx communities with many people generations removed from being immigrants (or whose families lived there before the area became part of the United States). Stakeholders also raised questions about why the African American community was not included in the discussion of subpopulations with particular histories of tension, although the National Initiative partners made that decision from the premise that issues of police-community trust with the African American community were so foundational and pervasive that they would necessarily be a focus of the reconciliation process in all participating sites. Finally, some respondents found the internal DOJ language of “subpopulation” off-putting or offensive, and worked to implement language for articulating the intentional focus on diverse areas of the community (such as the term “focus communities”) that was more conducive to respectful and effective public communication. The reconciliation work created new lines of communication with some subpopulations, which translated into some immediate changes in policy and practice. For example, a chief described what they heard in a listening session with the LGBTQIA+ community and how it led to action, saying, “There was a perception from that community that we did not work as hard when one of them was victimized … NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 77 From that session, we developed a policy, provided training on the policy, and we appointed a liaison to that community.” A police respondent from Stockton noted that the SPD had never met with LGBTQIA+ advocates in a dedicated session. Establishing this connection through a listening session led officers to work with a local community-based organization to learn how to use preferred pronouns and how those pronouns might not match government-issued identification. When I did my 30 years in law enforcement, we thought we knew the community, what they needed, and what was striking to me was that we didn’t know what we didn’t know. And what I mean to say is that the citizens in the community as a whole had these issues with us and we didn’t know and the reason we didn’t know we had them is because we never stopped down to listen to them. In the PJ training and the listening sessions, and I’ve been at all of them, they’ve been eye openers, and I just wish we had the listening session from 10– 15 years ago. Birmingham police respondent Though participant feedback regarding the reconciliation listening sessions was largely positive, there was a widespread belief that the value of individual listening sessions (and of the reconciliation process as a whole) depended on meaningful changes being made by the police department. Community participants wanted follow-up communication about whether and how listening sessions related to policy changes. Questions about whether and how the reconciliation process would lead to concrete changes dovetailed with concerns about the sustainability of the process beyond the end of the National Initiative grant, of which several community respondents were aware. There were concerns that momentum for change might dissipate after the National Initiative and that police practice would return to business as usual. In some sites, interviewees expressed that more capacity was needed to move policy reviews forward and effect change at the scale the National Initiative called for. The Pittsburgh Bureau of Police reviews each of its policies annually, but as one respondent noted, their policy unit consisted of a single person who was largely reviewing the policies to make sure they complied with accreditation requirements. Similarly, a Minneapolis respondent noted that only two people were working on policy in MPD, and that “that’s not really enough.” Where sites had substantial work to do defining and fully 78 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS staffing their policy review processes, community input in the policy process was limited or had yet to be established. Formal policy review processes could also be vulnerable to changes in police leadership, as when Birmingham and Gary experienced a change in chiefs. One way of addressing these concerns about sustainability was to publicly commit to sustaining the reconciliation work. Chief Jones did this in Stockton when he was quoted in the local media about the reconciliation process, saying, “We are never going to stop doing these.”70 Birmingham took the approach of engaging a community liaison to help facilitate and sustain the reconciliation process, although the city stopped engaging the liaison when National Initiative funding for the role ended in December 2018. (For more details, see NNSC [2019], a case study on the role of a community liaison in coordinating reconciliation work.) I’m enthused … by the connections that have been made. I don’t know how you measure that How do you measure them showing up and our young people not running? ‘Cause they would usually run. Stockton community respondent NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 79 Chapter 6: Conclusion The National Initiative spurred an array of changes and new practices intended to mitigate the mistrust between law enforcement and the most heavily policed communities, and to enhance legitimacy of police and the justice system. Delivering the National Initiative training continuum to every officer in the six National Initiative departments represented a major commitment of agency resources to infusing new concepts (and/or reinforcing or elevating important existing concepts) into policing practice. The reconciliation framework represented a substantial conceptual advance in the practice of improving relationships between police and communities, and implementing that framework developed new insights and innovations regarding how this process could look in practice. Moreover, police departments identified and made changes to their policies and structures in the interest of institutionalizing the change efforts they undertook through the National Initiative. The National Initiative training continuum began with the presentation of evidence in CPJ regarding the importance of police legitimacy, and applying procedural justice was the method for building legitimacy. The evaluation of the National Initiative was intended to assess whether the sites actually realized improvements in legitimacy. Surveys of residents of neighborhoods with the highest levels of crime and concentrated poverty found that perceptions of procedurally just treatment by police and the legitimacy of police improved from a pre-National Initiative baseline by a statistically significant margin, although improvements were not observed in all six cities. Similarly, perceptions of biased treatment by police decreased from baseline (Fontaine et al. 2019). As the largest demonstration project in the US to directly tackle the lack of trust between law enforcement and communities most impacted by crime, concentrated poverty , and violence, the National Initiative was a learning process. It generated many important lessons to guide future trust- building work. What follows are key takeaways from the implementation evaluation for other police departments and communities seeking to build trust: Police-community trust-building is a resource-intensive undertaking. The National Initiative partners invested considerable time in supporting the six sites in their efforts, including remote and on-site technical assistance and peer networks and exchanges. The DOJ supported the dedicated time of a site coordinator in each site. Moreover, site police departments and communities devoted large amounts of time and energy to National Initiative work. Leadership support and stability were important factors. Chiefs of police were critical players in all the National Initiative implementation components. However, turnover in chiefs is 80 NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT FINDINGS common, and changes in police leadership can disrupt organizational change and community relationships. Building and supporting champions for change bolsters trust building. Identifying credible messengers within each National Initiative police department not only helped transmit core procedural justice and implicit bias concepts to officers, it created subject matter experts on those concepts and champions for change who contributed to reconciliat ion and policy change work. Broadening police participation in reconciliation is important for sustainability. Chiefs were the main speakers for the police in the listening sessions, and participants appreciated hearing from them and considered this an important demonstration of leadership commitment. However, chiefs’ time is limited and chiefs changed during the initiative, so finding ways to delegate this role to conduct reconciliation activities at the requisite scale is important for sustainability. Police need to be willing to meet communities in their (safe) spaces and be mindful of how their uniforms can create distance. Engaging communities with difficult and traumatic histories with the police required being mindful of how to create spaces they would feel comfortable entering. This issue appeared to be most acute with youth. Holding meetings in neutral locations or locations of trusted organizations was one strategy for addressing this. Having officers attend meetings in street clothes rather than uniforms was another. Successful listening sessions required police to spend most of the sessions listening and to be nondefensive in their responses. Community participants who thought the listening sessions were positive frequently mentioned officers’ willingness to listen to what the community had to say (including angry and critical perspectives), and avoiding defensiveness was highly valued. Including the right people in the reconciliation listening sessions matters. Although community participants expressed mixed views about what constituted the “right people,” common groups included individuals critical of the police, members of particular communities e.g., survivors of domestic violence) and their advocates, and youth. Communities undertaking reconciliation work should be intentional when inviting participants and avoid the sense that participants are handpicked by police to avoid criticism. Including people with low levels of trust in the listening sessions will also make them more effective in identifying the causes of that mistrust. NATIONAL INITIATIVE IMPLEMENTATION ASSES SMENT F INDINGS 81 Follow-up communication with reconciliation participants is critical. Listening session participants, especially community members, were clear that they wanted the sessions to yield concrete changes in order to feel that the process was meaningful. The potential for building trust and improving relationships between police and communities can only be realized if changes (specifically those made in response to community input) are relayed to listening session participants. Policy change is a slow process that is particularly sensitive to leadership change. Although some individual policy changes occurred quickly in the sites, developing a policy review and revision process, particularly one involving meaningful community input, took substantial planning. Policy reviews and changes cannot move forward during leadership transitions because the department has to wait until the new chief indicates whether those changes are consistent with their vision. Trust-building also applies to relationships between experts and police. Police are skeptical of outside experts engaging with police departments in a culture-change process. Trust takes time to establish, and consistent communication is also important. Because of turnover, the National Initiative was more than a year into operation in the sites before this consistency was well- established. The positive trends in views of the police in the six cities’ most heavily policed neighborhoods suggest the National Initiative represents a promising intervention. Police professionals across sites often described the challenge of police-community trust-building with the axiom, “The communities that need us most trust us least.” This axiom captures the promise of an effort like the National Initiative: that meaningful trust-building (which must include the police becoming more trustworthy) can create new conditions that support safety and help people in all communities thrive. Though there is substantial work to be done in the six sites and across the country, the National Initiative offers promising ways of approaching that work. 82 APPENDIX Appendix A. National Initiative Site Timelines APPENDIX 83 84 APPENDIX APPENDIX 85 86 APPENDIX APPENDIX 87 88 NOTES Notes 1 Stakeholders in the National Initiative partner organizations and in the sites used the terms PJ1, PJ2, and PJ3 almost exclusively to refer to the three training components. Throughout this report, we employ the names conceptual procedural justice, tactical procedural justice, and implicit bias, respectively, for the three trainings. 2 The facilitator guide for TPJ is available at https://s3.trustandjustice.org/misc/PJ2__Facilitators_Guide_NI.pdf. 3 More information on the National Initiative, along with accompanying resources and tools for communities interested in engaging in similar efforts, can be found at www.trustandjustice.org. 4 Fort Worth designated a site coordinator but declined the DOJ funding to support the position. 5 Site coordinators participated in the monthly calls with the chiefs and the citywide calls. 6 “Community Facts,” US Census Bureau, https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml. 7 The FBI’s Uniform Crime Report identifies several cities that align with Birmingham in terms of population size and geography, including Columbus, Georgia, Shreveport, Louisiana, Hialeah, Florida, and Irving, Texas. In 2014, Birmingham’s population is estimated to have been 212,115 and its violent crime rate was 1,619 (per 100,000 people); Columbus had a population of 206,714 and a violent crime rate of 516; Shreveport had a population of 200,184 and a violent crime rate of 716; Hialeah had a population of 235,446 and a violent crime rate of 367; and Irving had a population of 231,708 and a violent crime rate of 271; “Crime in the US 2014,” FBI, https://ucr.fbi.gov/crime-in-the-u.s/2014/crime-in-the-u.s.-2014. 8 Carol Robinson, “Fifty Years Later, Birmingham Police Have Made Great Strides since Days of Dogs, Firehoses, Segregation.” Alabama Media Group, March 1, 2013, http://blog.al.com/spotnews/2013/03/fifty_years_later_birmingham_p.html. 9 Robinson, “Fifty Years Later.” 10 Deanna Boyd, “Black Officers Sue Fort Worth Ex-Police Chief,” Fort Worth Star-Telegram, January 13, 2015, http://www.star-telegram.com/news/local/crime/article6246291.html. 11 “Community Facts,” US Census Bureau. 12 “State-By-State and National Crime Estimates by Year(s),” Uniform Crime Reporting Statistics, http://www.ucrdatatool.gov/Search/Crime/State/RunCrimeStatebyState.cfm. 13 “Gary Counts Parcel Survey Web Tool,” City of Gary, accessed June 2016, http://garycounts.org/reports/. 14 US Department of Justice, “Former Gary, Indiana, Police Chief Sentenced for Federal Civil Rights Violation,” press release no. 09-357, April 16, 2009, https://www.justice.gov/opa/pr/former-gary-indiana-police-chief- sentenced-federal-civil-rights-violation. 15 Officer Westerfield was sitting in his car while responding to a domestic dispute call involving Blount . See also Elvia Malagon, “Trial Set in Homicide of Gary Patrolman Westerfield,” The Times of Northwestern Indiana, March 9, 2016, http://www.nwitimes.com/news/crime-and-court/trial-set-in-homicide-of-gary- patrolmanwesterfield/article_0a068632-3312-5d38-9027-0b020a1e9731.html. 16 Larry McKinley, “Guest Commentary: Gary Police Already Implementing Change,” The Times of Northwestern Indiana, April 7, 2016, http://www.nwitimes.com/news/opinion/columnists/guest-commentary/guest- commentary-gary-police-already-implementing-change/article_5571632e-4574-52c6-a539- 0f6fbaa1f7ac.html. NOTES 89 17 Gregory Tejeda, “Mayor, Council Member Say Gary Police, Fire to Get Pay Raise for 2016,” Chicago Tribune, August 5, 2015, https://www.chicagotribune.com/suburbs/post-tribune/ct-ptb-gary-police-raises-st-0806- 20150805-story.html. 18 “American Community Survey (ACS),” US Census Bureau, https://www.census.gov/programs- surveys/acs/news/data-releases.2014.html; “Overview,” Minnesota Compass, http://www.mncompass.org/immigration/overview. 19 John W. Miller, “America’s Most Literate Cities, 2014,” Central Connecticut State University, http://web.ccsu.edu/americasmostliteratecities/2014/default.asp. 20 “Minneapolis Public School District 4-Year Graduation Rate,” Minnesota Department of Education, http://rc.education.state.mn.us/#graduation/orgId--30001000000__groupType--district__graduationYearRate- 4__p--1. 21 “Shielded from Justice: Police Brutality and Accountability in the United States,” HRW (Human Rights Watch), 1998, https://www.hrw.org/legacy/reports98/police/uspo84.htm; “Officers Suspended After Putting Indians In Trunk,” Orlando Sentinel, April 24, 1993, http://articles.orlandosentinel.com/1993-04- 24/news/9304240580_1_trunk-police-department-bellecourt. 22 Reg Chapman, “Minneapolis Officer Who Allegedly Threatened to Break Teen’s Leg Fired,” CBS Local Minnesota, January 25, 2016, http://minnesota.cbslocal.com/2016/01/25/minneapolis-officer-who-allegedly-threatened- to-break-teens-leg-fired/; Brandt Williams, “Fong Lee's Family Angered by Verdict,” MPR News, May 28, 2009, http://www.mprnews.org/story/2009/05/28/fonglee_verdict. 23 Molly Priesmeyer, “94 Percent of Minneapolis Police Officers Live Outside the City of Minneapolis. Does It Matter?” MinnPost, December 3, 2014, https://www.minnpost.com/politics-policy/2014/12/94-percent- minneapolis-police-officers-live-outside-city-minneapolis-does-it. 24 Michael Krieger, “Two Weeks after Jordan Melee, Tensions Remain High,” Minnesota Daily, September 4, 2002, http://mndaily.com/2002/09/04/two-weeks-after-jordan-melee-tensions-remain-high; Brandt Williams and Meg Martin, “Jamar Clark Shooting, Protests Continue Strained History between Community, Police,” MPR News, December 4, 2015, http://www.mprnews.org/story/2015/12/04/jamarclark-shooting-history-north- minneapolis. 25 “Memorandum of Agreement,” Unity Community Mediation Team and the Minneapolis Police Department, December 3, 2003, http://www.minneapolismn.gov/www/groups/public/@council/documents/webcontent/convert_283328.pdf. 26 Williams and Martin, “Jamar Clark Shooting.” 27 “Engagement Summary, City of Minneapolis,” Office of Justice Programs Diagnostic Center, 2015, https://www.ojpdiagnosticcenter.org/wp-content/uploads/2019/05/Engagement-Summary-City-of- Minneapolis.pdf. 28 “Uniform Crime Reports - Crimes Reported to the Minneapolis Police Department, 2005-2015,” Minneapolis Police Department 2015, http://www.ci.minneapolis.mn.us/police/statistics/WCMS1P-104396. 29 Based on CODEFOR reports for weeks 832 (December 31, 2013–January 6, 2014) and 882 (December 16–22, 2014; YTD totals not available in week 883 report). “Suspicious Person” and “Suspicious Vehicle” are terms drawn directly from the reports; “Citywide - CODEFOR Statistics - Week 832,” Minneapolis Police Department, last updated April 30, 2018, http://www.minneapolismn.gov/police/statistics/crime-statistics_codefor_arrests; 90 NOTES Citywide - CODEFOR Statistics - Week 882,” Minneapolis Police Department, last updated April 30, 2018, http://www.minneapolismn.gov/police/statistics/crime-statistics_codefor_arrests. 30 Insufficient data was available on ethnicity to analyze the rates of arrests for low-level offenses among Latinx residents; “Picking up the Pieces: A Minneapolis Case Study,” ACLU, https://www.aclu.org/feature/picking- pieces. 31 “American Community Survey,” US Census Bureau. 32 Liz Navratil, “McLay’s Priority as Chief: ‘Bridging the Divide’,” Pittsburgh Post-Gazette, September 6, 2014, http://www.post-gazette.com/local/city/2014/09/05/Pittsburgh-s-new-police-chief-ready-to-get-to- work/stories/201409050311; Mark Nootbaar, “Pittsburgh Police Chief McLay Reflects Upon His First Year,” 90.5 WESA, September 15, 2015, http://wesa.fm/post/pittsburgh-police-chief-mclay-reflects-upon-his-first- year. 33 Megan Guza, “A Year in as Pittsburgh Police Chief, McLay Optimistic,” TribLive, September 16, 2015, http://triblive.com/news/allegheny/9021620-74/mclay-community-police; Nootbaar, “Pittsburgh Police Chief.” 34 Consent decree between the US District Court for the Western District of Pennsylvania and the Pittsburgh Bureau of Police, last updated August 6, 2015, https://www.justice.gov/crt/united-states-district-court- western-district-pennsylvania-united-states-america-plaintiff-v-0. 35 Deval L. Patrick, “City of Pittsburgh Investigative Findings Letter,” US Department of Justice, last updated August 6, 2015, https://www.justice.gov/crt/city-pittsburgh-investigative-findings-letter. 36 Davis and coauthors (2002) note that mandatory rotation has been one of the most controversial changes among officers and the union, who argue that it reduces officers’ ability to know the area they are patrolling and discourages bonds between officers and residents. 37 Consent decree between the US District Court for the Western District of Pennsylvania and the Pittsburgh Bureau of Police. 38 “Lawsuit over Pittsburgh Police’s Discriminatory Hiring Settled by ACLU and City,” ACLU, May 7, 2015, https://www.aclu.org/news/lawsuit-over-pittsburgh-polices-discriminatory-hiring-settled-aclu-and-city; Sadie Gurman, “ACLU Lawsuit Accuses Pittsburgh Police of Discriminating in Hiring,” Pittsburgh Post-Gazette, August 23, 2012, http://www.post-gazette.com/local/city/2012/08/23/ACLU-lawsuit-accuses-Pittsburgh-police-of- discrimination-in-hiring/stories/201208230369. 39 Margaret Harding, “Pittsburgh Police Officers Told to Stop Blaming ‘Peduto Head’ for Tickets,” TribLive, October 9, 2014, http://triblive.com/news/allegheny/6936877-74/peduto-officers-mayor; Margaret Harding and Bob Bauder, “Firefighters Accept Apology from Pittsburgh Mayor Peduto,” TribLive, May 20, 2014, http://triblive.com/news/allegheny/6146292-74/peduto-comments-employees. 40 Margaret Harding, “Among Contract Demands for Police: Less Overtime, Residences in City,” TribLive, August 11, 2014, http://triblive.com/news/allegheny/6597829-74/police-contract-comment. 41 Lexi Belculfine, “State Appellate Court Says Pittsburgh Police Officers Must Live in City,” Pittsburgh Post-Gazette, January 7, 2016, https://www.post-gazette.com/local/city/2016/01/07/State-appellate-court-says-Pittsburgh- police-officers-must-live-in-city/stories/201601070174. 42 “American Community Survey,” US Census Bureau. NOTES 91 43 Demographic data are available only for the most recent (2010) census, which measured Stockton’s population at 305,704. 2014 estimates are provided by the US Census Bureau for overall population (304,412) but not for different breakouts. 44 Kurt Badenhausen, “America’s Most Miserable Cities,” Forbes, February 2, 2011, http://www.forbes.com/2011/02/02/stockton-miami-cleveland-business-washington-miserable-cities.html; Kurt Badenhausen, “America’s Most Miserable Cities: Our Methodology,” Forbes, February 2, 2011, http://www.forbes.com/2011/02/02/stockton-miami-cleveland-business-washington-miserable-cities- methodology.html. 45 Jason Anderson, “Stockton Police Department Staffing Levels Continue to Rise,” The Record, June 30, 2016, http://www.recordnet.com/article/20160630/NEWS/160639965. 46 Tina Rosenberg, “A Strategy to Build Police-Citizen Trust,” New York Times, July 26, 2016, http://www.nytimes.com/2016/07/26/opinion/a-strategy-to-build-police-citizen-trust.html?_r=0. 47 John Rudolf, “Stockton’s Poor Mired in Violence after Police Cuts, Recession,” Huffington Post, March 19, 2012, http://www.huffingtonpost.com/2012/03/18/stockton-poor-poverty-crime-california_n_1346096.html. 48 “Personal Observations by SJTA President and Measure A Citizens’ Advisory Committee Member David Renison,” The City of Stockton, September 24, 2015. 49 Skogan, Van Craen, and Hennessey (2015) conducted a quasi-experimental evaluation of a daylong procedural justice training in the Chicago Police Department that showed that officers who were trained had greater support for procedural justice in the short term. However, the level of support was not sustained in the long term. The most recent study of procedural justice trainings involved a randomized control trial of 56 police recruits paired with a field training mentor, and it found some positive changes along attitudinal and behavioral measures (Antrobus, Thompson, and Ariel 2018). Owens and colleagues (2016) assigned officers to have meetings with their supervisors that focused on incorporating procedural justice contacts with civilians. Officers who participated in these meetings were less likely to be involved in incidents where force and arrest were used. 50 Less than one-quarter of all US police academies responded to the survey. 51 “A Rare Look inside NYPD’s Implicit Bias Training,” CBS News, August 17, 2018, https://www.cbsnews.com/news/nypd-rare-look-inside-police-implicit-bias-training/. 52 Devine and colleagues (2012) developed and implemented training for students on implicit bias. It focused on implicit bias as being a habit that could be overcome through awareness and concern and provided strategies for reducing bias. They found that students who participated in the trainings were significantly more likely to notice and label bias and even speak out against bias than those who did not participate in the training. More recently, Devine and colleagues (2017) administered a gender bias training based on the same premise to science, technology, engineering and math faculty at the University of Wisconsin-Madison and found that departments that participated in the training experienced significant changes in hiring patterns: in two years, the proportion of women faculty hired increased by nearly 50 percent in departments that participated in the training compared to those that did not. 53 The facilitator guide for CPJ is available at https://s3.trustandjustice.org/misc/PJ1__Facilitators_Guide_NI.pdf. 54 The facilitator guide for TPJ is available at https://s3.trustandjustice.org/misc/PJ2__Facilitators_Guide_NI.pdf. 55 The facilitator guide to the IB curriculum is available at https://s3.trustandjustice.org/misc/PJ3_Facilitators_Guide.pdf. 92 NOTES 56 Fort Worth and Pittsburgh had more trainers identified than they could send to the Chicago workshop, so they had to be caught up on the curriculum and incorporated into the training development. 57 All six sites referred to their police department National Initiative training teams as “PJ training teams,” inclusive of the trainers delivering the implicit bias component. 58 AASLH Contributing Author, “Civil Rights and a Civil Society,” American Association for State and Local History, February 1, 2017, https://aaslh.org/civil-rights-and-a-civil-society/. 59 Operation Ceasefire is Stockton’s application of the Group Violence Intervention strategy described on p. 57 of this report. 60 The attitude-behavior matching survey was designed to gauge respondents’ attitudes and psychological profiles and will be subsequently matched to individual-level performance data to understand how attitudes relate to behavior. 61 The National Initiative sites undertook many policy changes during the time that they were participating in the National Initiative with various local and national sources of inspiration, but this brief focuses on those changes made as a result of National Initiative participation. 62 City of Stockton, Police Department. 2017, “Rules and Regulations.” http://ww1.stocktonca.gov/Departments/Police/News-and-Information/Rules-and-Regulations 63 For more details on community input on deploying new technology in Stockton, see NNSC’s case study, Community Input on Unmanned Aerial Systems (Drones), Stockton Police Department.” 64 “34 Adults, 8 Juveniles Arrested as Jamar Clark Protest Shuts Down I-94,” Fox 9 News, November 16, 2015, http://www.fox9.com/news/50123489-story. 65 Associated Press, “Activists Demand Video of Jamar Clark Shooting, Shut Down Minneapolis Freeway,” Huffington Post, November 17, 2015. 66 KARE 11 Staff, “Jamar Clark Timeline: How Did We Get Here?” Kare 11, March 30, 2016, http://www.kare11.com/news/local/jamar-clark-timeline/99906515; Jennie Lissarrague, “Minneapolis Police Chief on Jamar Clark Protests: We're Dealing with Anarchists,” ABC 5 Eyewitness News, November 19, 2015, https://kstp.com/news/janee-harteau-jamar-clark-protesters-anarchists/3967984/. 67 Alex Baumhadt, Lindsey Bever, and Michael E. Miller, “Two Men Arrested in Shooting of Black Lives Matter Protesters in Minneapolis,” Washington Post, November 24, 2015, https://www.washingtonpost.com/news/morning-mix/wp/2015/11/24/five-people-shot-near-minneapolis- protest-cops-searching-for-3-white-male-suspects/. 68 Erin Golden, Paul Walsh, and David Chanen, “No Charges against Minneapolis Police Officers in Jamar Clark Shooting,” Star Tribune, March 31, 2016, http://www.startribune.com/no-charges-against-police-in-jamar-clark- shooting-death/373979481/#1. 69 US Department of Justice, “Federal Officials Decline Prosecution in the Death of Jamar Clark,” press release no. 16-634, June 1, 2016, https://www.justice.gov/opa/pr/federal-officials-decline-prosecution-death-jamar-clark. 70 Joe Goldeen, “Stockton Police Listening Tour Aimed at Building Trust,” Recordnet, April 28, 2017, https://www.recordnet.com/news/20170428/stockton-police-listening-tour-aimed-at-building-trust. REFERENCES 93 References Antrobus, Emma, Ian Thompson, and Barak Ariel. 2018. “Procedural Justice Training for Police Recruits: Results of a Randomized Control Trial.” Journal of Experimental Criminology 15 (1): 1–25. https://doi.org/10.1007/s11292- 018-9331-9. Aston, Elizabeth, Kath Murray, and Megan O’Neill. 2019. “Achieving Culture Change through Organizational Justice: The Case of Stop and Search in Scotland.” Criminology & Criminal Justice. https://doi.org/10.1177/1748895819839751. Berube, Alan, and Natalie Holmes. 2015. Some Cities Are Still More Unequal Than Others—an Update. Washington, DC: The Brookings Institution. Borum, Randy, Martha Williams Deane, Henry J. Steadman, and Joseph Morrissey. 1998. “Police Perspectives on Responding to Mentally Ill People in Crisis: Perceptions of Program Effectiveness.” Behavioral Sciences and the Law 16: 393–405. https://doi.org/10.1002/(SICI)1099-0798(199823)16:4%3C393::AID-BSL317%3E3.0.CO;2-4. Bradford, Ben, and Paul Quinton. 2014. “Self-Legitimacy, Police Culture and Support for Democratic Policing in an English Constabulary.” The British Journal of Criminology 54 (6): 1,023–46. https://doi.org/10.1093/bjc/azu053. Bradford, Ben, Paul Quinton, Andy Myhill, and Gillian Porter. 2013. “Why Do ‘the Law’ Comply? Procedural Justice, Group Identification and Officer Motivation in Police Organizations.” European Journal of Criminology 11 (1): 110–31. https://doi.org/10.1177%2F1477370813491898. Compton, Michael T., Masuma Bahora, Amy C. Watson, and Janet R. Oliva. 2008. “A Comprehensive Review of Extant Research on Crisis Intervention Team (CIT) Programs.” Journal of the American Academy of Psychiatry and the Law 36 (1): 47–55. Council on Crime and Justice and Institute on Race & Poverty. 2003. The Minnesota Racial Profiling Report. Minneapolis: Council on Crime and Justice and Institute on Race & Poverty. Davis, Robert C., Christopher W. Ortiz, Nicole J. Henderson, Joel Miller, and Michelle K. Massie. 2002. Turning Necessity Into Virtue: Pittsburgh’s Experience with a Federal Consent Decree. New York: Vera Institute of Justice. Devine, Patricia G., Patrick S. Forscher, Anthony J. Austin, and William T. L. Cox. 2012. “Long-Term Reduction in Implicit Race Bias: A Prejudice Habit-Breaking Intervention.” Journal of Experimental Social Psychology 48 (6): 1,267–78. https://dx.doi.org/10.1016%2Fj.jesp.2012.06.003. Devine, Patricia G., Patrick S. Forscher, William T. L. Cox, Anna Kaatz, Jennifer Sheridan, and Molly Carnes. 2017. A Gender Bias Habit-Breaking Intervention Led to Increased Hiring of Female Faculty in STEM Departments.” Journal of Experimental Social Psychology 73: 211–15. https://doi.org/10.1016/j.jesp.2017.07.002. DOJ (US Department of Justice). 2015. Investigation of the Ferguson Police Department. Washington, DC: US Department of Justice, Civil Rights Division. 2016. Investigation of the Baltimore City Police Department. Washington, DC: US Department of Justice, Civil Rights Division. Eith, Christine, and Matthew R. Durose. 2011. Contacts between Police and the Public. Washington, DC: US Department of Justice, Bureau of Justice Statistics. Fontaine, Jocelyn, Sino Esthappan, Nancy La Vigne, and Jesse Jannetta. 2019. “Views of the Police and Neighborhood Conditions: Evidence of Change in Six Cities Participating in the National Initiative for Building Community Trust and Justice.” Washington, DC: Urban Institute. Fort Worth CCPD (Crime Control and Prevention District). 2014. Crime Control and Prevention District: Summary of Funded Programs. Fort Worth, TX: FWPD. 94 REF ERENCES Glaser, Jack. 2014. Suspect Race: Causes and Consequences of Racial Profiling. New York: Oxford University Press. Hartney, Christopher, and Linh Vuong. 2009. Created Equal: Racial and Ethnic Disparities in the US Criminal Justice System. Oakland, CA: National Council on Crime and Delinquency. Hayner, Priscilla B. 2011. Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions, 2nd ed. New York: Routledge. Indiana State Police. 2013. “Gary Police Department Technical Assessment Team.” Indianapolis: Indiana State Police. Jones-Brown, Delores, Jaspreet Gill, and Jennifer Trone. 2010. Stop, Question & Frisk Policing Practices in New York City: A Primer. New York: The City University of New York, John Jay College of Criminal Justice, Center on Race, Crime, and Justice. Kneebone, Elizabeth, and Natalie Holmes. 2014. U.S. Concentrated Poverty in the Wake of the Recession. Washington, DC: Brookings Institution. La Vigne, Nancy, Jocelyn Fontaine, and Anamika Dwivedi. 2017. “How Do People in High-Crime, Low-Income Communities View the Police?” Washington, DC: Urban Institute. Lawrence, Daniel S., Nancy La Vigne, Jesse Jannetta, and Jocelyn Fontaine. 2019. Impact of the National Initiative for Building Community Trust and Justice on Police Administrative Outcomes. Washington, DC: Urban Institute. Logan, John R., and Brian J. Stults. 2011. The Persistence of Segregation in the Metropolis: New Findings from the 2010 Census. Providence, RI: Brown University. Nagin, Daniel S., and Cody W. Telep. 2017. “Procedural Justice and Legal Compliance.” Annual Review of Law and Social Science 13 (1): 5–28. https://doi.org/10.1146/annurev-lawsocsci-110316-113310. NNSC (National Network for Safe Communities). 2018a. “Case Study: Improving Police Service to Immigrant Communities in Pittsburgh.” New York: John Jay College of Criminal Justice, National Network for Safe Communities. 2018b. “Case Study: Applying Procedural Justice to Youth Engagements with School Resource Officers.” New York: John Jay College of Criminal Justice, National Network for Safe Communities. 2018c. “Case Study: Communicating the Connection between Trust Building and Violence Reduction.” New York: John Jay College of Criminal Justice, National Network for Safe Communities. 2019. “Case Study: Integrating Community Members into Reconciliation Listening Sessions in Birmingham.” New York: John Jay College of Criminal Justice, National Network for Safe Communities. OJP (Office of Justice Programs) Diagnostic Center. 2015. Diagnostic Analysis of Minneapolis Police Department, MN: Opportunities for Evidence-Based Technical Assistance. Presentation given to the Minneapolis Police Department, Minneapolis, January. Owens, Emily G., David Weisburd, Geoffrey Alpert, and Karen L. Amendola. 2016. Final Report: Promoting Officer Integrity through Early Engagements and Procedural Justice in the Seattle Police Department. Washington, DC: Police Foundation. Paluck, Elizabeth Levy, and Donald P. Green. 2009. “Prejudice Reduction: What Works? A Review and Assessment of Research and Practice.” Annual Review of Psychology 60: 339–67. https://doi.org/ https://doi.org/10.1146/annurev.psych.60.110707.163607. PBP (Pittsburgh Bureau of Police). 2016. Use of Force in the City of Pittsburgh: Initial Report, 2010 through 2015. Pittsburgh: City of Pittsburgh Bureau of Police. PERF (Police Executive Research Forum). 2013. Civil Rights Investigations of Local Police: Lessons Learned. Washington, DC: Police Executive Research Forum. REFERENCES 95 Pierson, Emma, Camelia Simoiu, Jan Overgoor, Sam Corbett-Davies, Vignesh Ramachandran, Cheryl Phillips, and Sharad Goel. 2017. “A Large-Scale Analysis of Racial Disparities in Police Stops across the United States.” Working Paper. Stanford, CA: Stanford University, Stanford Computational Policy Lab. Quattlebaum, Megan, Tracey Meares, and Tom Tyler. 2018. Principles of Procedurally Just Policing. New Haven, CT: The Justice Collaboratory at Yale Law School. Rosenbaum, Dennis, and Daniel S. Lawrence. 2012. Teaching Respectful Police-Citizen Encounters and Good Decision Making: Results of a Randomized Control Trial with Police Recruits. Washington, DC: National Institute of Justice, Department of Justice. Ross, Cody T. 2015. “A Multi-Level Bayesian Analysis of Racial Bias in Police Shootings at the County-Level in the United States, 2011–2014.” PLoS ONE 10 (11): e0141854. https://doi.org/10.1371/journal.pone.0141854. Seguino, Stephanie, and Nancy Brooks. 2018. A Deeper Dive into Racial Disparities in Policing in Vermont. Montpelier, VT: A Report Submitted to the Human Rights Commission of Vermont. Skogan, Wesley G., Maarten Van Craen, and Cari Hennessy. 2015. “Training Police for Procedural Justice.” Journal of Experimental Criminology 11 (3): 319–34. https://doi.org/10.1007/s11292-014-9223-6. Tyler, Tom R. 2004. “Enhancing Police Legitimacy.” The Annals of the American Academy of Political and Social Science 593 (1): 84–99. https://doi.org/10.1177/0002716203262627. Tyler, Tom R., Phillip Atiba Goff, and Robert J. MacCoun. 2015. “The Impact of Psychological Science on Policing in the United States: Procedural Justice, Legitimacy, and Effective Law Enforcement.” Psychological Science in the Public Interest 16 (3): 75–109. https://doi.org/10.1177%2F1529100615617791. 96 ABOUT THE AUTHORS About the Authors Jesse Jannetta is a senior policy fellow in the Urban Institute’s Justice Policy Center, where he leads research and technical assistance projects on reentry, justice system change efforts, community-based violence reduction strategies, and the practice of risk assessment in the criminal justice system. Sino Esthappan is a research analyst in the Urban Institute’s Justice Policy Center. His research focuses on policing, juvenile justice, and school discipline. Jocelyn Fontaine is a senior fellow in the Urban Institute’s Justice Policy Center. Her research incorporates quantitative and qualitative methodologies to evaluate the impact of community-based violence reduction and policing initiatives. Mathew Lynch is a former research associate in the Urban Institute’s Justice Policy Center. Nancy La Vigne is vice president for justice policy at the Urban Institute. She manages a staff of more than 50 scholars and conducts her own research on policing, criminal justice technologies, and reentry from incarceration. S TATEMENT OF I NDEPENDENCE The Urban Institute strives to meet the highest standards of integrity and quality in its research and analyses and in the evidence-based policy recommendations offered by its researchers and experts. We believe that operating consistent with the values of independence, rigor, and transparency is essential to maintaining those standards. As an organization, the Urban Institute does not take positions on issues, but it does empower and support its experts in sharing their own evidence-based views and policy recommendations that have been shaped by scholarship. Funders do not determine our research findings or the insights and recommendations of our experts. Urban scholars and experts are expected to be objective and follow the evidence wherever it may lead. 500 L’Enfant Plaza SW Washington, DC 20024 www.urban.org