LOOK AT THE ATTACHED DOCUMENT FOR BOTH ASSIGNMENTS.

168 Public Administration Review • March | April 2017

Public Administration Review,

Vol. 77, Iss. 2, pp. 168–178. © 2016 by

The American Society for Public Administration.

DOI: 10.1111/puar.12702.

Beyond Profiling: The Institutional Sources of Racial Disparities in Policing

Donald Haider-Markel is professor

and chair of political science at the

University of Kansas. His research and

teaching is focused on the representation

of interests in the policy process and the

dynamics between public opinion, political

behavior, and public policy.

E-mail: [email protected]

Steven Maynard-Moody is professor

in the School of Public Affairs and

Administration and director of the Institute

for Policy and Social Research at the

University of Kansas. He is coauthor, with

Charles Epp and Donald Haider-Markel, of

Pulled Over: How Police Stops Define Race and Citizenship (University of Chicago, 2014). With Michael Musheno, his current

research and writing extends the theoretical

frame first expressed in their book Cops, Teachers, Counselors: Stories from the Front Lines of Public Service (University of Michigan Press, 2003).

E-mail: [email protected]

Charles R. Epp is University

Distinguished Professor in the School

of Public Affairs and Administration at

the University of Kansas. He is author of

three books published by the University

of Chicago Press, including Making Rights Real (2009) on police reform and Pulled Over: How Police Stops Define Race and Citizenship (2014), with the coauthors of this article, on racial disparities in police

stops.

E-mail: [email protected]

Abstract : American policing faces a crisis of legitimacy. A key source of this crisis is a widespread police practice commonly endorsed by police leaders to fight crime. This is the investigatory stop, used to check out people who seem suspicious and to seize illegal drugs and guns and make arrests. Using data from an original scientific survey of drivers in the Kansas City metropolitan area, the authors show that racial disparities in police stops are concentrated in investigatory vehicle stops. In these stops, but not others, officers disproportionately stop African Americans and question and search them. The overwhelming majority of people stopped in this way are innocent, and the experience causes psychological harm and erodes trust in and cooperation with the police. Many of the most controversial police shootings during the past two years occurred in these stops. Reforming this practice is an essential step toward restoring trust in the police.

Practitioner Points • Although evidence of their effectiveness is not clear, investigatory police stops (commonly using minor

violations as a pretext for a more searching inquiry) are widely used by local police departments as a crime- fighting tactic.

• Most people stopped in investigatory stops are innocent, yet they are subjected to intrusive questioning (e.g., “Why are you in the neighborhood?”) and searches, leading to feelings of fear and of being “violated.”

• Overuse of investigatory police stops erodes trust in, and cooperation with, the police, especially among African Americans, who are especially likely to be stopped.

• There is insufficient oversight of the practice, as many investigatory stops yield no citation and so are not presently recorded or reported.

• To enable oversight of this practice, law enforcement agencies should require officers to record and report all stops they make, including the race and ethnicity of the driver and whether a warning or citation is issued; these data should be analyzed to check for patterns of racial disparity. Surveys of satisfaction with police services should include questions regarding residents’ experiences in police stops, including stops for minor violations.

Charles R. Epp Steven Maynard-Moody Donald Haider-Markel University of Kansas

Policing in the United States is in crisis, a “perfect storm” of popular protest and media coverage of egregious violations (Weitzer 2015 , 475). Since the protests in Ferguson, Missouri, over the shooting death of Michael Brown on August 9, 2014, protests have erupted in Baltimore, Charlotte, Cincinnati, Chicago, Cleveland, Los Angeles, Madison, Minneapolis-St. Paul, New York City, Oakland, St. Louis, Tulsa, and such smaller places as Hempstead, Texas; North Charleston, South Carolina; Pasco, Washington; and Stonewall, Mississippi. The relationship between the police and these communities, so essential to public safety and the rights and dignity of members of the public, is strained, if not broken. Nor is the problem isolated locally. A recent national survey found that 84 percent of African Americans believe that

blacks are treated less fairly by the police than whites (Stepler 2016 ).

Official mechanisms of police accountability have been mobilized. Eighteen individual officers were criminally indicted for police killings in 2015, the last year for which data are available, roughly triple the number in past years (Babwin 2015 ; Wing 2015 ). The U.S. Department of Justice has conducted formal investigations in Ferguson, Baltimore, and Chicago, on top of several others begun before the events in Ferguson. Several prominent police chiefs, including Baltimore police commissioner Anthony Batts, Chicago police superintendent Garry McCarthy, and San Francisco police chief Gregory P. Suhr, as well as a series of chiefs in Oakland, have been fired. The Barack Obama administration convened a high-level

Beyond Profi ling: The Institutional Sources of Racial Disparities in Policing 169

task force to suggest reforms of the problems of police violence and police-community conflict (President ’ s Task Force on 21st Century Policing 2015 ). In all, police departments, and the local governments they represent, are under mounting pressure to do something.

But what is the problem, and what should be done about it? Answers to these questions cluster into two broad types. One is that these tensions reflect the persistence of racist attitudes (whether intentional or implicit) among too many individual police officers. These individual-level biases, it is said, lead some officers to unfairly stop African American drivers or unnecessarily escalate the use of force when dealing with African American suspects. Locating the problem in individual beliefs or attitudes leads to reforms aimed at training officers away from acting on these beliefs or attitudes. This perspective is supported by a large body of psychological research and expressed in the efforts of major U.S. police professional associations and the Obama administration ’ s Task Force on 21st Century Policing to favor better training and control of individual frontline officers (President ’ s Task Force on 21st Century Policing 2015 ).

Although part of the problem surely is errant individual officers, this diagnosis and remedy misses how some of policing ’ s institutional structures and practices contribute to racial discrimination. Put simply, the problem is not errant individual officers alone but official policy and practice more broadly. In this article, we focus on these policies and practices and how they contribute to racial disparities in who is stopped, searched, and arrested. In doing so, we join a growing body of scholarship on the institutional sources of racial disparities (Ward and Rivera 2014 ). Thus, there is growing research on how institutionalized practices in education contribute to racial disparities in educational outcomes (see, e.g., Lucas 1999 , 2001 ). Health care research reveals that racial disparities in health outcomes grow not only from economic and geographic variations in access to health care but also from institutionalized practices related to diagnosis and treatment (see, e.g., Smedley, Stith, and Nelson 2002 ). We will argue that the current crisis in American policing is caused as much by standard police enforcement practices at the institutional level as by individual-level deviations from accepted standards.

Specifically, we argue that the investigatory police stop is a key source of racial disparities in police enforcement and a key source of African Americans’ distrust of the police. An investigatory stop is a stop of a driver (or pedestrian) that is made not to enforce traffic laws, vehicle codes, or the laws governing pedestrian activity but to check out people (or vehicles) who look suspicious. Is this driver carrying a gun or illegal drugs? What is he up to? Why is he in this neighborhood? Is there a warrant for his arrest? Stops for these purposes are made “proactively,” meaning they are made not on the basis of an observed violation but on a more inchoate suspicion: officers are to stop people to find out whether they are doing something wrong.

We will show that investigatory stops are a deep source of frustration and indignation among those who are stopped. We will

also argue that the practice of investigatory stops has become so frequently employed and so widespread that it is now a significant source of distrust in the police, especially among African Americans. We will further argue that recognizing these facts offers promise of a more effective reform program than the psychological explanation alone. The psychological approach to police reform views the problem as the “manner” by which officers carry out investigatory stops, meaning how politely they carry them out, not that they carry them out so frequently (Stuntz 2002 , 2174, reflecting the research of Tyler 2001 ; see esp. Tyler and Wakslak 2004 ). The psychological approach thus favors reforms aimed at training officers to improve their manners: to be more respectful in carrying out these stops. Although people who are stopped by the police appreciate being treated respectfully, our data reveal that people who are stopped for investigatory purposes resent the experience even if the officer is impeccably polite.

The problem, in other words, is the investigatory stop itself, not (only) officers who are carrying it out improperly; it is an institutional problem, not merely an individual problem. In recent decades, investigatory stops have become one of professional policing ’ s leading crime-fighting tactics, widely and frequently deployed in communities across the country. Yet, as we will show, this tactic harms the people who are stopped even when the officers

who do it follow the best training protocols. This harm and the distrust it produces falls disproportionately on racial minorities, as most of the people stopped in this way are African Americans and Latinos.

Presently, many police departments encourage their officers to make investigatory stops to fight crime yet carry out little oversight of whether these stops have the harmful effects we document here. As the Justice

Department ’ s report on the Baltimore Police Department observed, the likelihood of constitutional violations and harm to good police– community relations is greater where, as in Baltimore, a police department “does not collect reliable data on stops and searches, has no mechanism for identifying patterns or trends in its officers’ stops, searches, and arrests, and conducts little substantive review of officers’ reasons for taking particular enforcement actions” (2016, 26–27).

Thus, a final implication of our analysis is that evaluating police performance only in relation to crime rates and crime-clearance rates is shortsighted in that public trust in the police is affected as much by people ’ s experiences in police stops. These stops are often people ’ s most direct (and often viscerally powerful) experience of the police. Police departments should gather data on them, and city governments should exercise oversight of the practice. When city governments survey their constituents about their perceptions of police services (see Alfred Tat-Kai Ho and Wonhyuk Cho in this issue) without asking about experiences in police stops, they are likely to miss the single most substantial influence on these attitudes.

We proceed as follows: We first summarize the social scientific literature on investigatory stops. We then summarize our research

Although part of the problem surely is errant individual offi c- ers, this diagnosis and remedy misses how some of policing’s

institutional structures and practices contribute to racial

discrimination.

170 Public Administration Review • March | April 2017

on how, in the 1980s and 1990s, police leaders developed and refined the practice of investigatory stops and how this practice has been widely adopted by police departments as a key crime- fighting tactic. Then we summarize our data, reported elsewhere (Epp, Maynard-Moody, and Haider-Markel 2014 ), showing that investigatory stops are the key source of racial disparities in who is stopped by the police, who is subjected to intrusive questioning and searches during stops, and in people ’ s trust in the police.

This article builds on our analysis of these issues in Pulled Over: How Police Stops Define Race and Citizenship (2014). Here, we go beyond our analysis in that book by arguing that widespread investigatory stops of African Americans and Latinos contribute directly to the present crisis in American policing. Until the police accept this basic truth and rein in this practice, the current crisis is unlikely to be fully resolved.

Past Studies of Investigatory Police Stops Past studies of investigatory stops typically have focused on one or the other of two types of these stops: stops of pedestrians (“stop- and-frisks”) and stops of vehicles. Pedestrian stop-and-frisks were authorized by the U.S. Supreme Court in Terry v. Ohio (392 U.S. 1 [1968]) and are inherently investigatory, meaning that their essential purpose is to see whether the person who is stopped is engaged in criminality. As John A. Eterno, Christine S. Barrow, and Eli B. Silverman (in this issue) observe, these stops provoke widespread frustration among the people subjected to them. Although Floyd v. New York (959 F. Supp. 2d 540 [2013]) struck down that city ’ s widespread use of stop-and-frisk as racially discriminatory, Terry ’s central holding remains valid law—and stop-and-frisks are still widely used throughout the country.

If stop-and-frisks are inherently investigatory, vehicle stops, by contrast, include both standard stops to enforce traffic safety, which are generally not aimed at criminal investigation, and investigatory stops, which, like stop-and-frisks, are to check out the driver for criminal activity. Both types of vehicle stops are based on state statutes governing driving and the condition of vehicles. The U.S. Supreme Court in Whren v. U.S. (517 U.S. 806 [1996]) ruled that stops justified by minor traffic violations may be used as a pretext for a criminal investigation.

Although major U.S. police departments require officers to report stop-and-frisks as a distinct activity, they typically do not do so regarding investigatory vehicle stops, treating them instead as simply a vehicle stop like any other. Thus, while there are administrative data on pedestrian stop-and-frisks, only rarely are investigatory vehicle stops specifically identified in administrative data. This difference in reporting requirements may reflect the legal differences between pedestrian and vehicle stops.

These dissimilarities in law and administrative data have led to differences in how scholars have studied these two types of stops (of pedestrians versus of vehicles). With widely available administrative data on stop-and-frisks, there are excellent studies of racial bias in these stops (see, e.g., Gelman, Fagan, and Kiss 2007 ). By contrast, the vast majority of studies of racial bias in vehicle stops have not

conceptualized investigatory stops as a distinct type of vehicle stop (see, e.g., Lundman and Kaufman 2003 ; Petrocelli, Piquero, and Smith 2002 ; Smith and Petrocelli 2001 ; Tillyer and Engel 2013 ). As a consequence, most studies of vehicle stops implicitly view racial disparities in these stops not as a product of a police policy to encourage investigatory stops but as the result of individual officers’ discretionary choices about whom to stop. For example, Tillyer and Engel ( 2013 ) advance a “social conditioning” (i.e., psychological) model of the sources of racial disparities in vehicle stops.

These differences between the studies of pedestrian versus vehicle stops also have led to somewhat diverse estimates of the depth of racial disparities in the two types of stops. Substantial racial disparities are found in administrative data on pedestrian stop-and- frisks. Thus, Gelman, Fagan, and Kiss ( 2007 , 819–20), studying pedestrian stops in New York City and controlling for precinct-level

variations in racial demographics and crime rates, found that African Americans and Hispanics were stopped 2.5 times and 1.9 times, respectively, more than whites when the officer was investigating a violent crime and 1.8 and 1.6 times, respectively, more than whites when the officer was investigating a weapons violation.

By contrast, past studies of vehicle stops have reported racial disparities that, while significant, are in our view muddied by the difficulty of using administrative data to distinguish investigatory vehicle stops from other types of vehicle stops. For example, Tillyer and Engel ( 2013 ), studying all vehicle stops made by a large Ohio police agency, found that young black men were 1.3 times more likely than all other drivers to be issued a warning but somewhat less likely to be issued a citation. This difference is consistent with the hypothesis that stops of young black men are more likely to be made to investigate the driver for serious criminality (and then to release them with a warning for a minor violation) than to sanction him for a more serious traffic violation. Still, the degree of racial disparity is markedly less than found in studies of pedestrian stops in which the investigatory stop-and-frisk is sharply delineated in the administrative data. As we will show, clearly distinguishing investigatory and traffic safety stops yields estimates of the racial disparity in investigatory stops that are remarkably similar to those found in studies of pedestrian stop-and-frisks.

Our Contribution: Investigatory Stops as an Institutionalized Practice Our research goes beyond earlier studies by conceptualizing investigatory stops as a distinct type of vehicle stop, by developing a measure of its prevalence, and then by studying the effects of this type of stop on the attitudes of the people who are stopped in this way. Central to these advances is our conceptualization of investigatory vehicle stops as a distinct institutionalized practice. By this we mean that the investigatory stop has become a commonly structured police practice that, while not required by any specific official policy, is supported and legitimated by rules, training, and law and has spread widely to become a commonly accepted activity.

We developed the concept of an institutionalized practice to help make sense of why police departments widely deploy investigatory

Widespread investigatory stops of African Americans and

Latinos contribute directly to the present crisis in American

policing.

Beyond Profi ling: The Institutional Sources of Racial Disparities in Policing 171

vehicle stops even though there is no common federal mandate to do so and police departments are governed locally. The structure of U.S. policing is generally thought to ensure variation from department to department rather than commonality: the police are divided among tens of thousands of local agencies and dozens of state agencies. These myriad agencies operate under no common legal regime other than the constitutional law of criminal procedure. In spite of this common overarching legal regime, local political control is said to pull police departments in widely varying directions.

Nonetheless, police departments increasingly have adopted common organizational policies and practices (Epp 2009 ; Walker 1993 ), and we believe the investigatory stop is one such common practice. Our own research in the Kansas City metropolitan area documented the widespread deployment of investigatory stops among police agencies throughout that metropolitan area (Epp, Maynard-Moody, and Haider-Markel 2014 ). The police training manuals that we relied on in our book to illustrate this practice drew advice and guidance from police departments of virtually every size in all regions of the country (2014, 22–23). Recent ethnographic studies that follow up on our research observe investigatory stops in many jurisdictions in North Carolina (Coleman and Stuesse 2015 ) and in Nashville, Tennessee (Armenta 2016a , 2016b ). The recent official report on problems in the Chicago police department observes high numbers of investigatory stops, especially of African Americans, in that city (Police Accountability Task Force 2016 ). Our analysis of official publications of the International Association of Chiefs of Police documents widespread support among police agencies for using investigatory vehicle stops as a crime-fighting tactic (Epp, Maynard- Moody, and Haider-Markel 2014 , 36). In sum, there is considerable evidence of convergence among police agencies in training and directing their officers to carry out investigatory stops. We do not claim that every police department in the United States directs its officers to carry out investigatory stops. Rather, we claim that investigatory stops are widely deployed by many police departments of all sizes in all regions of the country and indeed are viewed as a “best practice,” by which we mean that police leaders view this practice as professionally vetted and effective.

Our concept of institutionalized practices draws on neoinstitutional theory to explain such a convergence among diverse police agencies. Neoinstitutional theory helps explain how separate organizations operating in a shared institutional field like policing often come to adopt common practices (DiMaggio and Powell 1983 ; Dobbin and Sutton 1998 ; Edelman 1992 ; March and Olsen 1984 ). The general lesson of these studies is that common institutionalized practices emerge from the sharing of ideas through professional networks rather than from official mandates. Over time, these practices take on “value beyond the technical requirements of the task at hand” (Selznick 1957 , 17). Put another way, institutionalized practices come to be viewed not only as effective but as professionally right, proper, and lawful. Through these processes of institutionalization, many diverse organizations in different legal jurisdictions come to have similar organizational structures and foster similar practices (DiMaggio and Powell 1983 ; Epp 2009 ; Garrow and Grusky 2012 ; Maynard-Moody and Musheno 2015 ).

Evidence of the widespread use of investigatory stops dates to the landmark studies of policing conducted in the wake of the 1960s urban riots. The Kerner Commission, appointed by President Lyndon B. Johnson, reported that the riots grew out of urban poverty and residents’ intense resentment of police stops and searches (National Advisory Commission on Civil Disorders 1968 ). The commission reported that investigatory police stops and searches were widely used in the cities that experienced riots and that resentment over these stops triggered many of the riots. Thus, the San Diego police had conducted from 20,000 to 40,000 stop-and-frisks per month, particularly in African American neighborhoods (Lohman and Misner 1966 , 127–34). A survey of citizens in 15 cities that had experienced riots found that while 6 percent of whites reported that police had frisked or searched them without a good reason, 22 percent of African Americans reported this experience (Campbell and Schuman 1969 , table IV).

The riots of the 1960s were, in part, uprisings against the practice of aggressive police stops. Many police departments certainly viewed them in this way and responded by backing away from their official commitment to aggressive stops. In the wake of that shift, studies of policing in urban neighborhoods in the 1970s described random police patrols—but without the widespread aggressive stops of the 1960s (Anderson 1978 ; Williams 1992 ).

Investigatory stops were reinvented after 1978 in a remarkable burst of activity. It began with James Q. Wilson and Barbara Boland ’ s call for police departments to revive what they termed “aggressive patrol”: “maximize[ing] the number of interventions in and observations of the community,” or, in other words, stopping and searching as many drivers as possible, “especially suspicious ones” (1978, 370–73). The federal government ’ s “war on drugs” brought this recommendation to fruition. Operation Pipeline, a key initiative of the Drug Enforcement Agency in the 1980s, trained state and local police nationwide to use pretextual traffic stops as a means to question and search large numbers of drivers for illegal drugs (Harris 2002 ; Webb 1999 ). Training films identified Hispanics and African Americans as more likely to be carrying drugs (Webb 1999 ). By the late 1990s, the DEA reportedly had trained some 27,000 state and local police officers in the practice of pretextual stops, and many of these officers went on to train others (Allen-Bell 1997 ; Harris 2002 , 48–51). In the 1990s, the police stop methods honed in the war on drugs

expanded into an effort to get guns off the street (Sherman and Rogan 1995 ; Sherman, Shaw, and Rogan 1995 ; Wilson 1994 ). In “hot spots” policing, another initiative of the 1990s, police were taught to focus their proactive stops on high-crime locations (Clarke and Weisburd 1994 ; Sherman, Buerger, and Gartin 1989 ; Sherman, Gartin,

and Buerger 1989 ; Weisburd and Mazerolle 2000 ). Around the same time, the National Highway Traffic Safety Administration (NHTSA) ( 1995 ) began encouraging police departments to use traffic stops as an all-purpose crime-control tool. NHTSA ’ s ongoing DDACTS program—Data-Driven Approaches to Crime and Traffic Safety—is the current instantiation of the agency ’ s long-standing support for using investigatory stops to suppress crime (NHTSA 2014 ).

Institutionalized practices come to be viewed not only as eff ec- tive but as professionally right,

proper, and lawful.

172 Public Administration Review • March | April 2017

The U.S. Supreme Court has lent constitutional legitimacy to the practice. Its landmark decision in Terry v. Ohio (1968) upheld stop-and-frisks of pedestrians. In 1996, in another landmark case, Whren v. U.S., the Court unanimously upheld the practice of pretextual vehicle stops made for the purpose of questioning a driver or searching a vehicle, so long as the officer has a legal justification for stopping the vehicle. The Court ’ s only caveat was that officers may not use a driver ’ s perceived race as the sole basis for making an investigatory stop, a prohibition that is technically important but practically meaningless as it is virtually impossible to enforce. In the wake of Whren, the Court extended police powers to stop and search vehicles in a series of key decisions ( Knowles v. Iowa, 525 U.S. 113 [1998]; Wyoming v. Houghton, 526 U.S. 295 [1999]; United States v. Arvizu, 534 U.S. 266 [2002]).

The legal and policy commitments favoring investigatory stops are translated into practice through training, and this has become systematic, enthusiastic, and detailed. In the authoritative and widely used Tactics for Criminal Patrol, Charles Remsberg teaches officers to use traffic stops “to maximize the number of citizen contacts in vehicle stops during each shift and, through specific investigative techniques, to explore the full arrest potential of each” (1995, 9). The text guides officers through a series of sequential steps that form what Remsberg ( 1995 , 9) called the “Criminal Patrol Pyramid,” summarized in table 1 . As a federal judge observed in Ligon v. City of New York (S.D.N.Y., Index No. 12 CIV 2274 [2013], 131), the New York police department ’ s training in how to conduct stop-and-frisks “has taught officers the following lesson: Stop and question first, develop reasonable suspicion later.”

In sum, investigatory stops are an institutionalized practice. Although not required by any common statute, many of the country ’ s myriad local police departments have come to favor this practice. It is taught and propagated by formal police training, favored in educational materials, extolled by national police professional associations and the shared norms of professional policing, and constitutionally legitimated by the Supreme Court. In fact, the shared professional commitment to this practice has deepened in recent decades. Once maligned as a trigger for the urban riots of the 1960s, by the 1990s the investigatory stop was hailed by police leaders as a best practice, as among professional policing ’ s most effective “crime-fighting tools” (Georges 2000 , 53).

Table 1 Charles Remsberg ’ s “Criminal Patrol Pyramid”

Steps in the Investigatory Stop

1. Develop suspicion (or, typically, merely curiosity) about a driver. 2. Discover a legal justification to stop the driver (typically this justification is

some minor violation of the traffic laws or vehicle code) and make the stop. 3. Decide, after making the stop, whether to seek to search the vehicle based

on the close observation of the vehicle and its visible contents and dialogue with the driver (and passengers). Officers use this dialogue to assess the truthfulness of the driver.

4. Search the vehicle (“usually by consent”). 5. Discover contraband or weapons. 6. Make an arrest. 7. Seek “bonus benefits” (forfeiture of vehicle, cash, etc.; information about

additional criminal offenses).

How Institutional Policy Interacts with Psychological Attitudes Our data, as we will describe later, reveal that racial disparities in who is stopped are concentrated in investigatory stops. There is no significant racial disparity in other types of stops. This pattern of no significant racial disparity in traffic safety stops but wide disparity in investigatory stops suggests that racial disparities in policing cannot be explained by individual officers’ choices alone; if so, we should find racial disparities in both kinds of police activity. The problem is thus an institutional phenomenon. Institutions are enacted by individuals, however (Maynard-Moody and Musheno 2012 ). The key question is how and why officers disproportionately select African American drivers for investigatory stops but not for traffic safety stops. Further, because the professional guidance regarding investigatory stops is racially neutral, and because few, if any, police departments are likely to direct their officers mainly to stop racial minorities in investigatory stops, we need to explain how a policy that is racially neutral in its formal terms nonetheless produces deep racial discrimination in practice.

Our answer is that officers, when directed by their superiors to stop people on the basis of suspicion of criminal activity, are likely to make choices about whom to stop on the basis of implicit racial stereotypes of black criminality. Department leaders are also likely to choose where to deploy investigatory stops on the basis of these stereotypes (Beckett, Nyrop, and Pfingst 2004 ; Beckett et al. 2005 ). (It is also likely that some officers are deliberately racist and intentionally choose to stop more African Americans, but our argument is not dependent on this assumption.) Although overt expression of racial bias arguably has declined since the 1960s (Bobo and Fox 2003 ), implicit negative stereotypes of African Americans persist. David Hamilton and Tina K. Trolier define “stereotype” as a “cognitive structure that contains the perceiver ’ s knowledge, beliefs, and expectations about human groups” (1986, 133). Patricia Devine observes that “ethnic attitudes and stereotypes are part of the social heritage of a society and no one can escape learning the prevailing attitudes and stereotypes assigned to major ethnic groups” (1989, 5). The stereotypes of African Americans are long-standing (Muhammad 2010 ) and generally negative: they are seen, especially by whites, as lazy and prone to violence and crime (Bobo 2004 ).

These individual attitudes on race matter. But our institutionalized practices thesis provides a more comprehensive explanation of racial disparities in police stops than the psychological explanation alone. That is because our institutionalized practices thesis predicts that racial disparities in police stops are likely to be concentrated in stops in which officers decide whom to stop on the basis of inchoate suspicion (rather than observed legal violations)—investigatory stops. In these settings, officers make quick decisions about whom to stop in a context of limited time and nonspecific information, encouraging what psychologists call “automatic” mental processes (Blair 2001 ; Devine 1986, 6; Greenwald et al. 2009 ). Unlike “controlled” thinking processes, such as deliberate problem solving, automatic processes are spontaneous and unintentional. They do not require conscious effort and are activated by environmental cues. Automatic or snap judgments are more likely to encourage the expression of embedded racial stereotypes (Blair 2001 ; Devine 1986, 6; Greenwald et al. 2009 ). Even individuals who express

Beyond Profi ling: The Institutional Sources of Racial Disparities in Policing 173

little overt prejudice—and indeed may consider racial prejudice morally wrong—may still respond on a gut level to racially biased stereotypes (Brodish and Devine 2005 , 58; Lepore and Brown 1997 ). For example, implicit stereotypes may contribute directly to such things as a police officer ’ s suspicion that a young black man seems out of place or poses a risk, to a decision to stop him to ask where he ’ s going, and so forth, even if the officer is professionally and morally committed to unbiased policing.

In sum, our theory of the institutionalized practice of investigatory stops predicts that racial disparities in police stops are likely to be concentrated in stops made on the basis of suspicion to proactively investigate the driver rather than on the basis of observed violation of traffic safety laws.

Methods and Data Our data were obtained from an original survey of drivers in the Kansas City metropolitan area. We designed this survey to overcome many of the methodological limitations of past studies of racial disparities in police stops, and so it is useful to summarize past studies’ methods before introducing our own. Most studies have relied on administrative data on traffic citations or stop-and-frisk reports. While these sources provide rich data about who is subjected to stop-and-frisks or issued traffic citations, they are plagued by what scholars call the “baseline problem,” or the absence of data necessary to establish a baseline against which to compare rates of stops or citations (Walker 2001 ; see also Ward 2002 ). As Gelman, Fagan, and Kiss observe, “causal claims about discrimination would require far more information about such baselines than the typical administrative (observational) data sets can supply” (2007, 815). For example, ideally, researchers would be able to compare stop rates among white and black drivers not only against the number of white and black drivers in the jurisdiction but also against a measure of how much time these different groups spend driving, how much they tend to violate traffic laws, and how much their vehicles have violations of the vehicle code. Yet researchers generally lack data on these measures and so are forced to rely for a baseline on simple totals of African Americans and whites in a jurisdiction.

Official administrative data on vehicle stops have two other key limitations. First, the vast majority of investigatory stops are never recorded because they end in no official citation or warning (but see Police Accountability Task Force 2016 for data on Chicago). Administrative data thus are likely to yield significant underestimates of the extent of racial disparities in stops. Second, administrative data contain no information about how the person who was stopped evaluated the stop or how the stop affected the individual ’ s attitude toward the police.

An alternative method, a survey of drivers, overcomes these limitations. The National Institute of Justice has conducted several national surveys of people ’ s contacts with the police, and the data provide a somewhat fuller set of control (baseline) variables (e.g., Engel and Calnon 2004 ; Lundman and Kaufman 2003 ). Still, even these otherwise excellent studies are confined to the rather limited range of data gathered in these official surveys.

Our study exploited the advantages of the survey method but asked a broader range of questions than in past surveys. Our questions were informed by focus groups that we conducted with officers from area departments. We asked drivers whether they had been stopped by the police in the past year and, if so, how often; the reason given for the stop; and whether a citation or warning was given and, if so, what violations were charged. Importantly, we asked about stops that yielded no citation as well as those that did. We also asked how both the driver and the officer spoke and acted during the stop and how the driver evaluated the stop and the drivers’ level of trust in the police and in local government. We asked drivers to report their race (both their self-identification and how their racial identity would be perceived by somebody meeting them for the first time), 1 gender, age, and income. We also asked a host of questions that provided data relevant to the development of a baseline (and to test alternative nonracial explanations of who is stopped), among them how much time the respondent spends driving per day and week, the condition of their vehicle, the type of clothes they typically wear, the extent to which they commit traffic violations, and so forth. This wide range of questions allowed us to test for racial bias in who is stopped and to assess the influence of police stops on people ’ s attitudes in a much fuller and more theoretically satisfying manner than in previous studies that did not include baseline measures and attitudinal indicators.

We surveyed a stratified random sample of the driving population in the Kansas City metropolitan area (spanning the Missouri–Kansas border), yielding responses from 2,329 drivers (for a full description of our survey methodology, see the methodological appendix in Epp, Maynard-Moody, and Haider-Markel 2014 ). We followed up the survey with in-depth interviews of a systematic subsample of drivers who reported being stopped; the interviews gathered stories of these stops in drivers’ own words and thus provided narrative data that reveal how police stops are carried out from start to finish.

Evidence of the Discriminatory Effects of Investigatory Stops Our analysis rests on empirically distinguishing investigatory vehicle stops from other stops. Our measure of investigatory stops is based on drivers’ responses to a survey question about the justification the officer gave for the stop upon approaching the driver. We classified investigatory stops as those in which the officer provided a de minimis justification for making the stop (e.g., driving two miles per hour over the posted limit) or provided no justification. All other stops we classify as traffic safety stops. These include, for example, stops made speeding (at seven or more miles over the limit), driving through a red light or stop sign, reckless driving, and suspicion of driving under the influence. (For a more detailed discussion of how we distinguished de minimis from other justifications, see Epp,

Maynard-Moody, and Haider-Markel 2014 , 59–64.)

Racial Disparities in Who Is Stopped and Searched Our basic observation is that racial disparities are concentrated in investigatory stops. When the police are enforcing traffic safety laws, such as stopping drivers for speeding at seven or more miles per hour over the limit

Whatever the psychological and cultural sources of racial dis-

parities are in police stops, these forces do not express themselves in traffi c safety enforcement, the most common street-level activ-

ity of police offi cers.

174 Public Administration Review • March | April 2017

or running a red light, they are not significantly more likely to stop black drivers (Epp, Maynard-Moody, and Haider-Markel 2014 , 64–66). Whatever the psychological and cultural sources of racial disparities are in police stops, these forces do not express themselves in traffic safety enforcement, the most common street-level activity of police officers .

When the police are carrying out investigatory stops, they are significantly more likely to stop black drivers. Controlling for a wide range of alternative explanations, African Americans were 2.7 times more likely than whites to experience an investigatory stop (see Epp, Maynard-Moody, and Haider-Markel 2014 , 65, table 3.1). Simply put, investigatory stops of vehicles especially target minority communities and people of color. Most whites, at least those over age 25, never experience this form of police stop (176). Put another way, in speeding stops, the most important influence on who is stopped is how fast you drive. In investigatory stops, the most important influence on who is stopped is not what you do but who you are: young black men are by far the most likely to be stopped.

These observations are based on multivariate logistic regression models that control for every conceivably significant alternative explanation for who is stopped: the driver ’ s age, gender, and other demographic characteristics; how much time the driver spends driving; the vehicle ’ s make, model, year, and other characteristics; the location of the stop; and so forth.

Likewise, inquisitive intrusions, for example, invasive questioning and searches, are not randomly scattered across all types of stops. They are concentrated in investigatory stops, and officers are much more likely to pursue these intrusions of African Americans than whites. Our data reveal that among drivers stopped for excessive speeding, African Americans are not significantly more likely to be questioned or searched. Intrusive questions and searches are concentrated in investigatory stops, and in these stops, African Americans are more than five times more likely than whites to be searched (Epp, Maynard-Moody, and Haider-Markel 2014 , 81–84, 103–5). This disparity in searches occurs even though officers are considerably less likely to discover contraband or weapons in searches of African Americans than whites (105).

Harms to the Individuals Who Are Stopped Our analysis revealed as well that these two types of stops have profoundly different psychological and social consequences for individuals and communities. The consequences of investigatory stops are especially harmful, akin to the costs to individual, family, and community caused by racial discrimination, as described by Feagin and McKinney ( 2002 ). Our data show that while white drivers’ experience of predictable, reasonable traffic safety stops contributes to their perception of their equal status in a rule-governed democracy, African Americans’ common experience of investigatory stops contributes to their perception that they are not regarded by the police as full and equal members of society (Epp, Maynard-Moody, and Haider-Markel 2014 , 134–51). Investigatory stops, we find, are significantly more likely to foster the perception that the police are “out to get people like me,” and they render people significantly more likely to carefully select the clothes they wear and avoid driving in

some areas for fear of how the police might treat them (145–46). For example, two-thirds of African American men under age 30 (compared with 40 percent of white men this age) say they avoid driving in some areas of the city for fear of how the police may treat them. Even for those ages 40 to 49, nearly half of African Americans of both sexes,

compared with 17 percent of white men and 8 percent of white women, say they avoid some areas for fear of the police. Stunningly, our data show that investigatory stops—but not traffic safety stops—even make people less comfortable with calling the police for help. (An earlier study observed this effect for police stops in general but, unlike our study, was not

able to distinguish the effects of traffic safety and investigatory stops; see Gibson et al. 2010 ). These are deep psychological costs. They contribute to the ongoing recreation of racial segregation.

Another cost of investigatory stops is deep, palpable fear of the police. Investigatory stops, our data demonstrate, sometimes produce dramatic escalations of tension between the officer and the person stopped (Epp, Maynard-Moody, and Haider-Markel 2014 , 74–92). These tensions reflect the fact that people who are subjected to intrusive investigations on the pretext of a minor violation, such as a burned-out taillight, think that this is unfair and unjust, and they fear what the officer may do next. Thus, if an officer can handcuff you and search your car when all you did was forget to fix that little light, what is to stop him from doing far worse? Where is this stop heading? People subjected to these stops told us of feeling deeply fearful and resentful, of feeling “violated” (1). Police officers, too, recognize that the people who are stopped and searched under these circumstances are resentful and tense, and so officers are also tense and ready to act forcefully to keep control of the interaction. Our interviews reveal that both police officers and African Americans who experience investigatory stops realize that escalations are possible and try hard to avoid it.

Sometimes tensions escalate, however, confirming African American drivers’ fear that these stops may spiral out of control into violence. Billy, a black man, told us he once asked an officer, “Why are you stopping me, because I ’ m a poor man in a ragged car?” Billy said the officer responded by yelling, “Show me your hands, show me your hands.” “He had his pistol pulled out,” Billy said, “and I didn ’ t even turn my head, but I could see it out of the peripheral vision of my eye, its barrel, and he pointed it right at my head, you know. That was pretty frightening” (Epp, Maynard-Moody, and Haider- Markel 2014 , 90–91). Even when the driver is let go from such an experience with no physical harm, psychological pain endures.

Sometimes these inevitable tensions escalate to violence. In fact, several of the most controversial police shootings since that of Michael Brown in Ferguson, Missouri, have occurred in precisely such a context. These include the shootings of Walter Scott in North Charleston, South Carolina; Samuel DuBose in Cincinnati, Ohio; and Philando Castile in Minnesota, which occurred in vehicle stops that bear all of the markings of investigatory stops and in which tensions between the officer and the driver escalated (Apuzzo and Williams 2015 ; Pérez-Peña 2015 ; Stahl 2016 ). The arrest of Sandra Bland, who died in a Texas jail, likewise was made in the midst of escalating tensions in such a stop (Lai et al. 2015 ).

Investigatory stops—but not traffi c safety stops—even make people less comfortable with calling the police for help.

Beyond Profi ling: The Institutional Sources of Racial Disparities in Policing 175

Harms to Community and Society The costs of investigatory stops include social and community costs (Feagin and McKinney 2002 ). We have already noted how investigatory stops contribute to the ongoing racial segregation of our urban areas. Another key cost is to the police themselves: our data reveal that investigatory stops powerfully erode trust in the police among the people subjected to these stops. We measured trust in the police with a range of standard questions concerning this trust as well as questions unique to our survey. We combined these questions into an index of trust in the police ( α = .91). Our data reveal that the experience of an investigatory stop significantly erodes trust in the police; in fact, the experience of such a stop erodes trust as much as being stopped by an officer acting with extreme rudeness and disrespect (put more technically, the effect of an investigatory stop is as great as experiencing an average stop carried out by an officer in the top 25th percentile on our scale of officer rudeness) (Epp, Maynard- Moody, and Haider-Markel 2014 , 143–44). The experience of a traffic safety stop, by contrast, has no significant effect on trust in the police. This contrast is consistent with our thesis that the investigatory stop is a distinct practice with distinct, deleterious effects, including the loss of legitimacy of the police themselves. This erosion in trust in the police surely is part of the explanation for the current crisis in policing.

One other social effect deserves brief mention. Our data suggest that the investigatory stop shapes people ’ s perceptions of race and racial identities. “Race,” of course, is a cultural construction (see, e.g., American Anthropological Association 1998 )—and it is also increasingly clear that some government policies and practices contribute to racial divisions and identities (Gomez 2010 ). We find that the experience of an investigatory stop among African Americans is significantly associated with an increase in people ’ s identification with, and loyalty to, African Americans as a group. Strikingly, we find that the experience of an investigatory stop has the opposite effect among whites, reducing their identification with the white race. These results remain statistically significant when controlling for a wide range of known influences on racial identification. The effects of investigatory stops on racial identification are consistent with a growing body of evidence (see, e.g., Lerman and Weaver 2014 ) that the experience of intrusive police stops is a defining element of what is called a sense of “linked fate” among African Americans, or the belief that being black in American society is a powerful source of shared experiences of discrimination and solidarity (Dawson 1994 ).

Discussion Investigatory stops are an institutionally supported practice that causes harm both to the people who are stopped and to police legitimacy. This practice sacrifices the liberty and dignity of large numbers of innocent people who are disproportionately racial minorities in pursuit of a small number who are dangerous or carrying contraband (Harcourt 2007 ). The innocent victims of this practice rightly feel that their status as free and equal members of society has been diminished, and they resent this implication and the police for enforcing this message. The victims of this practice have learned from their experiences to distrust the police. Many now are demanding fundamental police reforms.

Leaders of professional policing occasionally have acknowledged these deeper truths and recognize how investigatory stops contribute

to the crisis they face. Thus, two police leaders (Milazzo and Hansen 1999 ) observed,

Despite the constitutionality of the practice [of investigatory stops], motorists understand that they are being stopped for a different reason than the one provided and are angered by the inference needed to make the stop. The resulting anger could escalate hostility and risk greater physical danger for both the officer and motorist. The anger continues after the stop and is shared within the motorist ’ s peer group. As more anecdotal stories circulate about such stops, the long-term effect on race relations must be balanced against the short-term effect on drug enforcement. Police managers may legitimately weigh race relations as more important than the potential arrest of a drug user or street level dealer.

The Police Executives Research Forum (PERF) similarly has acknowledged the harms caused by investigatory stops. A PERF- published report on racially biased policing urged departments to train officers away from using race as a key factor in deciding whom to stop and search. Yet the report also acknowledged that the “conventional police strategy,” that “numerous stops, searches, citations, and arrests will yield reductions in crime, disorder and accidents,” is itself a key source of racial disparities in stops (Fridell et al. 2001 , 93). The report concluded, “oftentimes, intensive criminal and traffic enforcement falls short of the desired effects, and instead, only worsens the relationship between police and the minority community” (93).

These observations, made more than 15 years ago, now seem prescient. In the brief time since the police shooting of Michael Brown in Ferguson, a broad movement for police reform has revealed how little many members of our society trust that the police are engaged in fair and impartial law enforcement. This perception is not simply a legacy of the past. Our data demonstrate that investigatory police stops contribute to it daily: investigatory stops fuel distrust in the police.

The problem, put simply, is not only the occasional errant, racist officer who violates police policy and training but the “best practice” of investigatory stops itself. This practice is deeply institutionalized in contemporary policing. The police have not acted alone in developing this policy, however. Support for the so-called war on crime is long-standing and shared across the political spectrum (Simon 2007 ). Americans ask our police departments to reduce crime. Professional policing responded by developing “proactive” crime-fighting tactics, chief among them the investigatory stop. To those favoring investigatory stops, the experience of such a stop seems a minor inconvenience for the innocent, a fair price to pay for getting drugs and guns off the streets.

The evidence that investigatory stops help fight crime is surprisingly weak. Although an old study suggested that investigatory stops might help reduce crime (Sherman and Rogan 1995 ; Sherman, Shaw, and Rogan 1995 ), more recent studies conclude that there is no clear evidence that investigatory stops help reduce crime (see, e.g., Cohen and Ludwig 2003 ; Koper and Mayo-Wilson 2006 ; McGarrell et al. 2001 ; Zimring 2011 ). Supporters of investigatory stops seem unaware of this fact.

176 Public Administration Review • March | April 2017

Nor do they recognize the considerable costs of investigatory stops to individual dignity, the community and police–community relations. Scattershot investigatory stops, even if ostensibly focused on those who look suspicious, mainly target innocent community residents for highly invasive intrusions. Many people who experience these stops describe them as akin to “incarcerations.” Like people everywhere, they appreciate police protection, but they also value fair treatment and freedom from arbitrary intrusion: they, like everyone, resent being treated like criminals. Stopped in this way, people begin to avoid driving in some areas for fear of how the police may treat them; they begin to fear calling the police for help and therefore decrease the likelihood of cooperation with police. Investigatory stops thus erode individuals’ basic liberties and dignity. These harms extend to the police and society itself. Investigatory stops erode the trust in police so necessary for police legitimacy. They also diminish community empowerment (Rosenbaum 2006 ) and what Robert Sampson and Stephen Raudenbush (1999) call “collective efficacy,” the fundamental community social asset that helps to reduce crime. Investigatory stops may yield some seizures of drugs and guns and a few arrests of criminals, but the long-term cost in the erosion of trust between the police and community reduces the ability of the police and residents to prevent and solve crime over the longer term.

Consider the social and community costs of erosion in the willingness of people to call the police for help, one of the effects of experiencing an investigatory police stop. What options are open to somebody who has been the victim of a crime but who distrusts the police and so is unwilling to call them for help? Such a person may decide to do nothing and simply try to go on with life as best he (or she) can after the crime. Or he may decide to take matters into his own hands and seek retribution (Gibson et al. 2010 ). Both represent fundamental violations of the equal protection of the law. They also contribute to more crime, further erosion of lawful norms, and continuing degradation of the quality of life in some communities. In these ways, investigatory police stops are deeply counterproductive.

What, then, is to be done? Ending investigatory stops, except when compellingly justified by clear and particular cause, would contribute directly to improved police–community relations. The leading approaches to police reform in the wake of Ferguson, as we noted earlier, focus on better training and control of individual officers, especially training to improve officer respectfulness. Our data confirm that drivers evaluate stops more positively the more the officer is polite and respectful. But our data also demonstrate that this is hardly enough. African American drivers commonly experience investigatory stops and resent the practice; our data show that they evaluate these stops more negatively than traffic safety stops no matter how polite and respectful is the officer. In fact, among these drivers the corrosive effect of an investigatory stop (compared with a traffic safety stop) on a driver ’ s evaluation of the stop is as substantial as the effect of being stopped by an officer whose behavior is at the 25th percentile (compared with the 75th percentile) on our scale of officer politeness/impoliteness (Epp, Maynard-Moody, and Haider-Markel 2014 , 129–33). Put simply, investigatory stops are as poisonous to drivers’ evaluations of the stop as being

stopped by a deeply rude and disrespectful officer. Strikingly, African Americans who experience a traffic safety stop by a polite officer evaluate the experience no differently than a white driver in a similar stop. The larger lesson is clear: to improve African Americans’ perceptions of the police, rein in the practice of investigatory stops.

Steps in this direction would include requiring officers to record and report the race, gender, and age of all persons stopped, including stops in which no warning or citation is issued. Currently, many investigatory stops are not recorded. This lack of reporting limits the capacity of command-level officers to monitor the actions of their officers and look for patterns of racial disparity. It also limits oversight by city managers and elected officials. Additionally, surveys of city residents regarding satisfaction with police services should include questions about experiences in police stops, including stops for minor violations.

Conclusion When confronted with new evidence of racial disparities in police stops, most discussions turn, yet again, to denials that it is a widespread problem or claims that it is the product of a few bad apples. In fact, as our research demonstrates, the unjust disparities in police stops are widespread and the result of deeply institutionalized practices that are supported by law and policy and refined and engrained by training and socialization. Although formally color blind, investigatory stops encourage the expression of powerful, deeply ingrained negative stereotypes of black and brown Americans. This police practice thus actively recreates our society ’ s racial divide. It reconstructs the meaning of rights, citizenship, and race in contemporary America, contributing to racial inequality rather than equality (Ward and Rivera 2014 ).

Reforming investigatory stops will not be easy. The practice is viewed by many in policing not as a problem but as a good thing. In many places, doing investigatory stops is part of what it means to be a good

cop. Still, although institutionalized practices resist change they are not unchangeable. If the professional leaders of our cities and police departments are willing to learn about the immense costs of the investigatory stop practice, and are willing to confront the fact that it is this practice, and not only the actions of a few bad apples, that has so eroded public support for the police, then the door may open to real institutional reform. Police agencies should replace this “scattershot, stop-lots-of-people-in-the hope-of-catching-a- few” practice (Forman and Stutz 2012 ) with police–community collaboration. The leadership of American policing should recognize that widespread investigatory stops are counterproductive and morally repugnant. They should define this institutionalized practice as unprofessional and train and evaluate officers in alternative, community-based approaches.

Acknowledgments This article is based on research that was made possible by a grant from the National Science Foundation (SES-0214199). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Investigatory police stops are deeply counterproductive.

Beyond Profi ling: The Institutional Sources of Racial Disparities in Policing 177

Note 1 . For each of these questions, we allowed drivers to select more than one of the

standard ethnoracial categories. Virtually all drivers selected one primary racial category, and there was extremely little difference between drivers’ self- identification and their estimate of another person ’ s first impression of their racial identity. In the present analysis, we use the self-identification measure.

References Allen-Bell , Angela Anita . 1997 . The Birth of the Crime: Driving while Black

(DWB) . Southern University Law Review 25 : 195 – 225 . American Anthropological Association . 1998 . AAA Statement on Race. http://www.

aaanet.org/stmts/racepp.htm [accessed November 11, 2016]. Anderson , Elijah . 1978 . A Place on the Corner . Chicago : University of Chicago Press . Apuzzo , Matt , and Timothy Williams . 2015 . Citizen ’ s Videos Raise Questions on

Police Claims . New York Times , April 9 . Armenta , Amada . 2016a . Between Public Service and Social Control: Policing

Dilemmas in the Era of Immigration Enforcement . Social Problems 63 ( 1 ): 111 – 26 .

______ . 2016b . Racializing Crimmigration: Structural Racism, Colorblindness and the Institutional Production of Immigrant Criminality. Sociology of Race & Ethnicity. Published electronically on May 17. doi:10.1177/2332649216648714.

Babwin , Don . 2015 . Number of Police Charged with Killings up in 2015 . Boston Globe, December 4 , 2015 .

Beckett , Katherine , Kris Nyrop , and Lori Pfingst . 2004 . Race, Drugs and Policing: Understanding Disparities in Drug Delivery Arrests . Criminology 44 ( 1 ): 105 – 38 .

Beckett , Katherine , Kris Nyrop , Lori Pfingst , and Melissa Bowen . 2005 . Drug Use, Drug Possession Arrests, and the Question of Race: Lessons from Seattle . Social Problems 52 ( 3 ): 419 – 41 .

Blair , Irene V. 2001 . Implicit Stereotypes and Prejudice . In Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition , edited by Gordon B. Moskowitz , 359 – 74 . Mahwah, NJ : Lawrence Erlbaum .

Bobo , Lawrence D. 2004 . Inequalities That Endure? Racial Ideology, American Politics, and the Peculiar Role of Social Science Changing the Terrain of Race and Ethnicity . In The Changing Terrain of Race and Ethnicity , edited by Maria Krysan and Amanda E. Lewis , 13 – 42 . New York : Russell Sage Foundation .

Bobo , Lawrence D. , and Cybelle Fox . 2003 . Race, Racism, and Discrimination: Bridging Problems, Methods, and Theory in Social Psychology Research . Social Psychology Quarterly 66 ( 4 ): 319 – 32 .

Brodish , Amanda B. , and Patricia G. Devine . 2005 . The Dynamics of Prejudice, Stereotyping, and Intergroup Relations: Intrapersonal and Interpersonal Processes . Social Psychological Review 71 ( 1 ): 54 – 70 .

Campbell , Angus , and Howard Schuman . 1969 . Racial Attitudes in Fifteen American Cities . In Supplemental Studies for the National Advisory Commission on Civil Disorders . Washington, DC : U.S. Government Printing Office .

Clarke , Ronald V. , and David Weisburd . 1994 . Diffusion of Crime Control Benefits: Observations on the Reverse of Displacement . Crime Prevention Studies 2 : 165 – 84 .

Cohen , Jacqueline , and Jens Ludwig . 2003 . Policing Crime Guns . In Evaluating Gun Policy: Effects on Crime and Violence , edited by Jens Ludwig , and Philip J. Cook , 217 – 50 . Washington, DC : Brookings Institution Press .

Coleman , Mat , and Angela Stuesse . 2015 . The Disappearing State and the Quasi- Event of Immigration Control . Antipode 48 ( 3 ): 524 – 43 .

Dawson , Michael . 1994 . Behind the Mule: Race and Class in African-American Politics . Princeton, NJ : Princeton University Press .

Devine , Patricia G. 1989 . Stereotypes and Prejudice: Their Automatic and Controlled Components . Journal of Personality and Social Psychology 56 ( 1 ): 5 – 18 .

DiMaggio , Paul J. , and Walter W. Powell . 1983 . The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields . American Sociological Review 48 ( 2 ): 147 – 60 .

Dobbin , Frank , and John R. Sutton . 1998 . The Strength of a Weak State: The Employment Rights Revolution and the Rise of Human Resources Management Divisions . American Journal of Sociology 104 ( 2 ): 441 – 76 .

Edelman , Lauren B. 1992 . Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law . American Journal of Sociology 97 ( 6 ): 1531 – 76 .

Engel , Robin Shepard , and Jennifer M. Calnon . 2004 . Examining the Influence of Drivers’ Characteristics during Traffic Stops with Police: Results from a National Survey . Justice Quarterly 21 ( 1 ): 49 – 90 .

Epp , Charles R. 2009 . Making Rights Real: Activists, Bureaucrats, and the Creation of the Legalistic State . Chicago : University of Chicago Press .

Epp , Charles R. , Steven Maynard-Moody , and Donald Haider-Markel . 2014 . Pulled Over: How Police Stops Define Race and Citizenship . Chicago : University of Chicago Press .

Feagin , Joe R. , and Karyn D. McKinney . 2002 . The Many Costs of Racism . Lanham, MD : Rowman & Littlefield .

Forman , James , Jr., and Trevor Stutz . 2012 . Beyond Stop and Frisk . New York Times , April 19 .

Fridell , Lori , Robert Lunney , Drew Diamond , and Bruce Kubu , with Michael Scott and Colleen Laing . 2001 . Racially Biased Policing: A Principled Response . Washington, DC : Police Executive Research Forum .

Garrow , Eve , and Oscar Grusky . 2012 . Institutional Logic and Street-Level Discretion: The Case of HIV Test Counseling . Journal of Public Administration Research and Theory 23 ( 1 ): 103 – 31 .

Gelman , Andrew , Jeffrey A. Fagan , and Alex Kiss . 2007 . An Analysis of the New York City Police Department ’ s “Stop-and-Frisk” Policy in the Context of Claims of Racial Bias . Journal of the American Statistical Association 102 ( 479 ): 813 – 23 .

Georges , William P . 2000 . Traffic Safety Strategies Yield Tangible Benefits. Police Chief, July, 53.

Gibson , Chris L. , Samuel Walker , Wesley G. Jennings , and J. Mitchell Miller . 2010 . The Impact of Traffic Stops on Calling the Police for Help . Criminal Justice Policy Review 21 ( 2 ): 139 – 59 .

Gomez , Laura E. 2010 . Understanding Law and Race as Mutually Constitutive: An Invitation to Explore an Emerging Field . Annual Review of Law and Social Science 6 : 487 – 505 .

Greenwald , Anthony G. , T. Andrew Poehlman , Eric Luis Uhlmann , and Mahzarin R. Banaji . 2009 . Understanding and Using the Implicit Association Test: III. Meta-Analysis of Predictive Validity . Journal of Personality and Social Psychology 97 ( 1 ): 17 – 41 .

Hamilton , David L. , and Tina K. Trolier . 1986 . Stereotypes and Stereotyping: An Overview of the Cognitive Approach . In Prejudice, Discrimination, and Racism , edited by John F. Dovidio , and Samuel L. Gaertner , 127 – 63 . Orlando, FL : Academic Press .

Harcourt , Bernard E. 2007 . Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age . Chicago : University of Chicago Press .

Harris , David A. 2002 . Profiles in Injustice: Why Racial Profiling Cannot Work . New York : New Press .

Koper , Christopher S. , and Evan Mayo-Wilson . 2006 . Police Crackdowns on Illegal Gun Carrying: A Systematic Review of Their Impact Gun Crime . Journal of Experimental Criminology 2 ( 2 ): 227 – 61 .

Lai , K. K. Rebecca , Haeyoun Park , Larry Buchanan , and Wilson Andrews . 2015 . Assessing the Legality of Sandra Bland ’ s Arrest. New York Times, July 22.

Lepore , Lorella , and Rupert Brown . 1997 . Category and Stereotype Activation: Is Prejudice Inevitable? Journal of Personality and Social Psychology 72 ( 2 ): 275 – 87 .

Lerman , Amy E. , and Vesla M. Weaver . 2014 . Arresting Citizenship: The Democratic Consequences of American Crime Control . Chicago : University of Chicago Press .

Lohman , Joseph D. , and Gordon E. Misner . 1966 . The Police and the Community: The Dynamics of Their Relationship in a Changing Society . Washington, DC : U.S. Government Printing Office .

Lucas , Samuel R. 1999 . Tracking Inequality: Stratification and Mobility in American High Schools . New York : Teachers College Press .

178 Public Administration Review • March | April 2017

______. 2001 . Effectively Maintained Inequality: Education Transitions, Track Mobility, and Social Background Effects . American Journal of Sociology 106 ( 6 ): 1642 – 90 .

Lundman , Richard J. , and Robert L. Kaufman . 2003 . Driving While Black: Effects of Race, Ethnicity, and Gender on Citizen Self-Reports of Traffic Stops and Police Actions . Criminology 41 ( 1 ): 195 – 220 .

March , James G. , and Johan P. Olsen . 1984 . The New Institutionalism: Organizational Factors in Political Life . American Political Science Review 78 ( 3 ): 734 – 49 .

Maynard-Moody , Steven , and Michael Musheno . 2012 . Social Equities and Inequities in Practice: Street-Level Workers as Agents and Pragmatists . Special issue, Public Administration Review 72 : S16 – 23 .

______. 2015 . Embedded Agency and Inhabited Institutions: Accounting for Patterns in Frontline Worker Judgment. Paper presented at the International Conference on Public Policy, Milan, Italy, July 1–4.

McGarrell , Edmund F. , Steven Chermak , Alexander Weiss , and Jeremy Wilson . 2001 . Reducing Firearms Violence through Directed Police Patrol . Criminology and Public Policy 1 ( 1 ): 119 – 48 .

Milazzo , Carl , and Ron Hansen . 1999 . Race Relations in Police Operations: A Legal and Ethical Perspective . Paper presented at the 106th Annual Conference, International Association of Chiefs of Police , Charlotte, NC , October 30– November 3.

Muhammad , Khalil Gabran . 2010 . The Condemnation of Blackness: Race, Crime, and the Making of Modern Urban America . Cambridge, MA : Harvard University Press .

National Advisory Commission on Civil Disorders (Kerner Commission) . 1968 . Report of the National Advisory Commission on Civil Disorders . Washington, DC : U.S. Government Printing Office .

National Highway Traffic Safety Administration (NHTSA) . 1995 . Traffic Enforcement: Saving Lives and Combating Crime . Washington, DC : NHTSA .

______ . 2014 . Data Driven Approaches to Crime and Traffic Safety (DDACTS): Operational Guidelines. https://www.nhtsa.gov/staticfiles/nti/ddacts/811185_ DDACTS_OpGuidelines.pdf [accessed November 11, 2016].

Pérez-Peña , Richard . 2015 . University of Cincinnati Officer Indicted in Fatal Shooting of Driver. New York Times, July 29.

Petrocelli , Matthew , Alex R. Piquero , and Michael R. Smith . 2002 . Conflict Theory and Racial Profiling: An Empirical Analysis of Police Traffic Stop Data . Journal of Criminal Justice 31 ( 1 ): 1 – 11 .

Police Accountability Task Force . 2016 . Recommendations for Reform: Restoring Trust between the Police and the Communities They Serve . Chicago : Police Accountability Task Force . https://chicagopatf.org/wp-content/ uploads/2016/04/PATF_Final_Report_4_13_16-1.pdf [accessed November 11, 2016].

President ’ s Task Force on 21st Century Policing . 2015 . Final Report of the President ’ s Task Force on 21st Century Policing. https://cops.usdoj.gov/pdf/taskforce/ taskforce_finalreport.pdf [accessed November 11, 2016].

Remsberg , Charles . 1995 . Tactics for Criminal Patrol: Vehicle Stops, Drug Discovery and Officer Survival . Northbrook, IL : Calibre Press .

Rosenbaum , Dennis P. 2006 . The Limits of Hot Spots Policing . In Police Innovation: Contrasting Perspectives , edited by David Weisburd , and Anthony A. Braga , 245 – 63 . New York : Cambridge University Press .

Sampson , Robert J. , and Stephen W. Raudenbush . 1999 . Systematic Social Observation of Public Spaces: A New Look at Disorder in Urban Neighborhoods . American Journal of Sociology 105 ( 3 ): 603 – 51 .

Selznick , Philip . 1957 . Leadership in Administration: A Sociological Interpretation . Berkeley : University of California Press .

Sherman , Lawrence , Michael Buerger , and Patrick Gartin . 1989 . Repeat Call Address Policing: The Minneapolis RECAP Experiment . Final Report to the National Institute of Justice. Washington, DC : Crime Control Institute .

Sherman , Lawrence W. , Patrick R. Gartin , and Michael E. Buerger . 1989 . Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place . Criminology 27 ( 1 ): 27 – 55 .

Sherman , Lawrence W. , and Dennis P. Rogan . 1995 . Effects of Gun Seizures on Gun Violence: “Hot Spots” Patrol in Kansas City . Justice Quarterly 12 ( 4 ): 673 – 93 .

Sherman , Lawrence W. , James W. Shaw , and Dennis P. Rogan . 1995 . The Kansas City Gun Experiment . Washington, DC : National Institute of Justice .

Simon , Jonathan . 2007 . Governing through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear . New York : Oxford University Press .

Smedley , Brian D. , Adrienne Y. Stith , and Alan R. Nelson , eds. 2002 . Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care . Washington, DC : National Academies Press .

Smith , Michael R. , and Matthew Petrocelli . 2001 . Racial Profiling? A Multivariate Analysis of Police Traffic Stop Data . Police Quarterly 4 ( 1 ): 4 – 27 .

Stahl , Brandon . 2016 . Officer ’ s Lawyer: Castile Pulled Over Because He Matched Robbery Suspect. Star-Tribune (Minneapolis-St. Paul, MN), July 11.

Stepler , Renee . 2016 . 5 Key Takeaways about Views of Race and Inequality in America. Pew Research Center, June 27. http://www.pewresearch.org/fact-tank/2016/06/27/ key-takeaways-race-and-inequality/ [accessed November 11, 2016] .

Stuntz , William J. 2002 . Local Policing after the Terror . Yale Law Journal 111 ( 8 ): 2137 – 94 .

Tillyer , Rob , and Robin Shepard Engel . 2013 . The Impact of Drivers’ Race, Gender, and Age During Traffic Stops: Assessing Interaction Terms and the Social Conditioning Model . Crime & Delinquency 59 ( 3 ): 369 – 95 .

Tyler , Tom R. 2001 . Public Trust and Confidence in Legal Authorities: What Do Majority and Minority Group Members Want from Law and Legal Institutions? Behavioral Sciences and the Law 19 ( 2 ): 215 – 35 .

Tyler , Tom R. , and Cheryl J. Wakslak . 2004 . Profiling and Legitimacy of the Police: Procedural Justice, Attributions of Motive, and Acceptance of Social Authority . Criminology 42 ( 2 ): 253 – 81 .

U.S. Department of Justice . 2016 . Investigation of the Baltimore City Police Department. https://www.justice.gov/opa/file/883366/download [accessed November 11, 2016].

Walker , Samuel . 1993 . Taming the System . New York : Oxford University Press . ______ . 2001 . Searching for the Denominator: Problems with Police Traffic Stop Data

and an Early Warning System Solution . Justice Research and Policy 3 ( 1 ): 63 – 95 . Ward , James D. 2002 . Race, Ethnicity, and Law Enforcement Profiling: Implications

for Public Policy . Public Administration Review 62 ( 6 ): 726 – 35 . Ward , James D. , and Mario A. Rivera . 2014 . Institutional Racism, Organizations and

Public Policy . New York : Peter Lang . Webb , Gary . 1999 . DWB . Esquire, April 1 : 118 – 28 . Weisburd , David , and Lorraine Green Mazerolle . 2000 . Crime and Disorder in Drug

Hot Spots: Implications for Theory and Practice in Policing . Police Quarterly 3 ( 3 ): 331 – 49 .

Weitzer , Ronald . 2015 . American Policing under Fire: Crisis and Reform . Society 52 ( 5 ): 475 – 80 .

Williams , Terry . 1992 . Crackhouse: Notes from the End of the Line . Reading, MA : Addison-Wesley .

Wilson , James Q. 1994 . Just Take Away Their Guns . New York Times Magazine ( March 20 ).

Wilson , James Q. , and Barbara Boland . 1978 . The Effect of Police on Crime . Law & Society Review 12 ( 3 ): 367 – 90 .

Wing , Nick . 2015 . 16 Numbers That Explain Why Police Reform Became an Even Bigger Story in 2015. Huffington Post, December 29. http://www.huffingtonpost.com/entry/police-reform-numbers- 2015_5672e150e4b0688701dc7a54 [accessed November 11, 2016].

Zimring , Franklin E. 2011 . The City That Became Safe: New York ’ s Lessons for Urban Crime and Its Control . New York : Oxford University Press .