Columbia Law School Columbia Law School

Scholarship Archive Scholarship Archive

Faculty Scholarship Faculty Publications

2018

Do the Ends Justify the Means? Policing and Rights Tradeoffs in Do the Ends Justify the Means? Policing and Rights Tradeoffs in

New York City New York City

Amanda Geller abg2108@columbia.edu

Jeffrey Fagan Columbia Law School, jfagan@law.columbia.edu

Tom R. Tyler tom.tyler@nyu.edu

Follow this and additional works at: https://scholarship.law.columbia.edu/faculty_scholarship

Part of the Criminal Law Commons, Criminal Procedure Commons, and the Law Enforcement and

Corrections Commons

Recommended Citation Recommended Citation Amanda Geller, Jeffrey Fagan & Tom R. Tyler, Do the Ends Justify the Means? Policing and Rights Tradeoffs in New York City, COLUMBIA PUBLIC LAW RESEARCH PAPER NO. 14-581 (2018). Available at: https://scholarship.law.columbia.edu/faculty_scholarship/2092

This Working Paper is brought to you for free and open access by the Faculty Publications at Scholarship Archive. It has been accepted for inclusion in Faculty Scholarship by an authorized administrator of Scholarship Archive. For more information, please contact scholarshiparchive@law.columbia.edu.

Electronic copy available at: https://ssrn.com/abstract=3122655

Do the Ends Justify the Means? Policing and Rights Tradeoffs in New York City§

Amanda Geller*

Jeffrey Fagan†

Tom R. Tyler‡

ABSTRACT

Policing has become an integral component of urban life. New models of proactive policing

create a double-edged sword for communities with strong police presence. While the new

policing creates conditions that may deter and prevent crime, close surveillance and frequent

intrusive police-citizen contacts have strained police-community relations. The burdens of the

new policing often fall on communities with high proportions of African American and Latino

residents, yet the returns to crime control are small and the risks of intrusive, impersonal,

aggressive non-productive interactions are high. As part of the proffered tradeoff, citizens are

often asked to view and accept these invasive tactics as a necessary means to the ends of reduced

crime and improved public safety. This paper examines the degree to which urban residents’

show a willingness to engage in a “rights tradeoffs” and sacrifice their civil liberties to maintain

public safety. Using a telephone phone survey of 960 New York City residents, we find little

openness to rights tradeoffs tied to perceived neighborhood danger. However, respondents who

see the police as legitimate and effective in producing safety are more likely to support such

tradeoffs. The results suggest that trust in the police can give them wide berth to infringe on

civil liberties in the interest of crime control, regardless of local crime conditions, the

abrasiveness of police contact, and the extent and type of the intrusions on privacy and liberty.

§ This research was generously supported by the Open Society Foundation through Grant No. 20034571.

The authors thank Chelsea Davis for outstanding research assistance. All opinions and errors are solely

those of the authors. An earlier version of the paper was presented at the Conference on Empirical Legal

Studies, Cornell Law School, October 2017. * Associate Clinical Professor, Department of Sociology, New York University † Isidor and Seville Sulzbacher Professor of Law and Professor of Epidemiology, Columbia University) ‡ Macklin Professor of Law and Professor of Psychology, Yale University

Electronic copy available at: https://ssrn.com/abstract=3122655

Do the Ends Justify the Means? Policing and Rights Tradeoffs in New York City

Amanda Geller*

Jeffrey Fagan†

Tom R. Tyler‡

I. BACKGROUND

For people in cities, and especially in neighborhoods with elevated crime rates, the police

are more than the face of the state: the presence of the police reflects both the political belonging

of residents in the liberal democratic state, but also the state’s controlling interest in maintaining

security (Soss and Weaver, 2017). Police produce order and safety, providing an amenity to the

everyday lives of citizens. In turn, police are an integral component of urban life, with the

potential to drastically improve or safeguard the well-being of city residents (Skogan & Frydl,

2004). Yet, that view of police is now contested, and varies according to the extent and quality

of citizen contacts with police. In an era when proactive and often aggressive policing is widely

used (Weisburd and Majmandur, 2017), particularly in urban areas and their neighborhoods with

the highest crime rates, policing may be seen as a dis-amenity and a negative feature of

neighborhood ecology. When policing is experienced as unnecessary, abusive and degrading,

citizens may distance themselves from the police and develop perspectives that reject the

underlying norms and moral authority of the police (Tyler and Fagan, 2008; Hough et al., 2010;

Jackson and Gau, 2016). Whether policing is seen as an amenity or disamenity varies both with

the policing tactics and the lived experiences of residents who interact with police.

Police are indeed present in the everyday lives of people, and their place in the social

ecology of urban life, is borne out by estimates from both population and survey data. The 2011

* Associate Clinical Professor, Department of Sociology, New York University † Isidor and Seville Sulzbacher Professor of Law and Professor of Epidemiology, Columbia University) ‡ Macklin Professor of Law and Professor of Psychology, Yale University

Electronic copy available at: https://ssrn.com/abstract=3122655

Police-Public Contacts survey shows that that more than a quarter of the adult population (26%)

had one or more involuntary contacts with the police in the preceding year (Langton & Durose,

2013). A survey of Chicago public school students found that approximately half had been

stopped, questioned, and “told off or told to move on” by ninth or tenth grade (Shedd, 2015).

Evidence from the National Longitudinal Survey of Youth shows that by age 18, the cumulative

arrest prevalence rates1 range from 30.2-41.4% (Brame et al., 2012, 2014). Data from a new

population-based survey of urban teens indicate that more than 20% report personal experience

with the police, and 75% report knowing someone with police contact or witnessing a police stop

(Geller, 2018).

At the same time, police are an important deterrent of crime (Durlauf & Nagin, 2011),

though the empirical link between the presence of police and specific policing tactics is uncertain

and contingent on the policing tactics applied. Beyond crime control, police also provide

important services that go well beyond their primary crime control mission. Police are often first

responders in a variety of circumstances, from fire and safety to responding to emotionally

disturbed persons or family crises, tasks that take an emotional and psychological toll on police

(Marmar et al., 2006; Rando et al., 2015). Police can also play a key role in the legal

socialization of residents of the cities (Justice and Meares, 2014; Fagan & Tyler, 2005; Tyler &

Fagan, 2008; Tyler, Fagan, & Geller, 2014). Residents who have positive experiences with the

police, in which they feel they have been treated fairly and respectfully, tend to place more trust

in the police as an institution, which in turn, has the potential to increase police-public

cooperation, and encourage citizen participation in the production of public safety (Meares, 2014

Fagan & Tyler, 2005; Kirk & Papachristos, 2011). Yet here too is a seeming contradiction:

whatever the returns of policing to public safety, there may also be incursions on dignity and

privacy that for many can produce mental health symptoms including trauma, anxiety and stigma

(Geller, Fagan, Tyler and Link, 2014; Sewell and Jefferson, 2016; Sewell, Jefferson, and Lee,

2016).

Several empirical studies have shown how these tensions shape the responses of citizens

to the police and their assessments of police legitimacy (Meares, 2002; Trinkner, Jackson and

Tyler, 2017). These studies focus on evaluations of the police, while other research examines

how these evaluations influence law-related behaviors including cooperation and compliance

(Tyler and Fagan, 2008; Berg et al., 2016). In this paper, we extend this question to a broader

normative assessment: how citizens balance or comparatively weight their privacy and liberty

rights relative to the promise of security by police, and whether these rights are fungible in an

exchange for their safety (Whitman, 2003); Bayley, 2002; Sargeant, 2017; Solove, 2007; Popkin

et al., 1995). Courts often weigh the balance between liberty and security in deciding the limits

of police authority to contravene privacy rights in the interest of public safety (Pratt v. Chicago

Housing Authority, 1994; Gutierrez et al. v. City of East Chicago, 2016). Courts have invoked

the necessity of deference to law enforcement interests to ensure public safety in navigating the

balance of privacy rights and safety and tolerating incursions of Fourth Amendment privacy

rights, Fourteenth Amendment due process rights, Fourteenth Amendment Equal Protection

rights (c.f., United States. v. Brignoni-Ponce, 1975; Whren v. United States, 1996; Johnson,

2009.).

Our question arises from the experiences of over five million New Yorkers during the era

of intensive stop-and-frisk activity from 2004-2014 who were stopped by the New York City

Police Department (Floyd v City of New York, 2013). Citizens often were detained on the street

at low levels of suspicion; the constitutionality of the majority of those stops was questioned and

ultimately rejected by the Floyd court. The Floyd court also cited sharp racial imbalances in the

this practice (Fagan et al., 2010; Fagan, 2017a). Most of these stops entailed temporary

detentions, some involved force by police, but rarely increased public safety by resulting in

arrests or seizures of contraband (White and Fradella, 2016; Fagan, 2010, 2012). The stops often

were degrading and verbally if not physically abusive (Howell, 2009). Persons stopped by the

police elsewhere reported similar interactions, including physical assaults, degrading language,

unnecessary or even gratuitous use of force, yet rarely resulting in arrests or recovery of guns or

other contraband (Brunson and Weitzer, 2010; Harcourt, 2004; Epp, Maynard-Moody and

Haider-Markel, 2014; Jain, 2015; Rios, 2011; Bowers, 2014). In addition to legitimacy costs to

the police, young persons stopped by the police exhibited symptoms of post-traumatic stress,

anxiety and stigma (Geller, Fagan, Tyler and Link, 2014).

During this era, police officials and supporters have suggested that low-level police

contact may deter crime (Smith & Purtell, 2008; Weisburd, Wooditch, Weisburd, & Yang,

2016), and as such, is instrumental to maintaining public safety in high-crime neighborhoods

(MacDonald, 2012). Other studies have questioned either the practical or statistical significance

of the marginal policing effects on crime (Rosenfeld and Fornango, 2014; MacDonald, Fagan

and Geller, 2016), and suggest that the costs to legitimacy and cooperation among the most

intensively policed citizens may offset these minor gains in public safety (Weisburd et al., 2016).

Other studies suggest that this form of proactive policing can produce large reductions in crime,

but only when the majority of police contacts are conducted at a high bar of constitutional

compliance (Fagan, 2016). Despite these realities, residents of heavily policed areas have been

encouraged to consider police intrusion a “fact of urban life”, and to consider any associated

imposition on civil liberties a small price to pay in exchange for the resulting improvements to

public safety (Williams, Fromer, Fagbenle, & Stein, 2012).

Explicit in this message was the notion that at least a partial tradeoff of equal protection,

privacy rights and dignity for increased safety was a bargain that was required by one’s presence.

The increased surveillance and police activity in non-white areas was incorporated into this

algebra with the explanation that minority citizens were also the primary beneficiaries of police

activity (c.f., MacDonald, 2010); Zimring, 2011).

Here, we examine the extent to which residents in fact were willing to make the

exchanges proposed by the proponents of the aggressive policing policies of the past 20 years.

We analyze evidence from New York City, but the experiences of these residents are not unique

to experiences in other cities across the U.S. (Fagan and Richman, 2017). We ask how policing

affects the political and normative choices of citizenship among people living in communities

where both crime and police are constant presences, each potentially intruding on their privacy,

dignity and security.

A. Proactive Policing

Despite the importance of policing as a determinant of public safety, the role of the police

in American cities has been an increasingly contentious topic over the past two decades

(Harmon, 2012). Many cities have shifted to “proactive” (Kubrin, Messner, Deanne, McGeever,

& Stucky, 2010; Weisburd and Majmandur, 2017) and “Broken Windows” (Heymann, 2000;

Kelling & Coles, 1996; Wilson & Kelling, 1982) policing models, which emphasize the active

engagement of residents at low levels of suspicion, in order to detect imminent criminal activity.

In this “new policing”, police use investigative stops, arrests, and citations to disrupt low-level

disorder, or other circumstances interpreted as indicia that crime is afoot (Kelling & Coles, 1996;

Kubrin et al., 2010; Wilson & Kelling, 1982). However, the vast majority of these encounters do

not result in arrest, citation, or the seizure of contraband (Fagan, 2010; Fagan, Conyers, & Ayres,

2014), suggesting that most people stopped by the police have done nothing wrong (Herbert,

2010; Morrow, White, and Fradella, 2017). Such encounters are also characterized by stark racial

disparities in many cities, with the greatest burden of police contact falling on racial and ethnic

minority residents (American Civil Liberties Union of Illinois, 2015; Ayres & Borowsky, 2008;

Bailey et al. v. City of Philadelphia et al., 2011; Fagan, et al., 2016) and neighborhoods (Fagan

and Davies, 2000; Fagan, Davies, & Carliss, 2012; Fagan, Geller, Davies, & West, 2010; Geller

& Fagan, 2010).

These low-level, racially disparate encounters, coupled with high-profile incidents of

police violence (Hagelskamp, Friedman, Rizzolo, Rinehart, & Schleifer, 2015; Krieger, Chen,

Waterman, Kiang, & Feldman, 2015) and misconduct (Rayman, 2010a, 2010b, 2010c), have the

potential to compromise perceptions of police legitimacy (Fagan & Tyler, 2005; Tyler & Fagan,

2008), and threaten community health. (Geller, Fagan, Tyler, & Link, 2014; Sewell & Jefferson,

2016; Sewell, Jefferson, & Lee, 2016).

B. Public Safety and Support for Civil Liberties

Throughout American history, citizens have expressed unequivocal support for civil

liberties (e.g., the rights to free speech, assembly, and due process) and related democratic

principles (McClosky & Brill, 1983; Sniderman, Fletcher, Russell, & Tetlock, 1996); however,

this support has repeatedly been compromised in the face of apparent threat (ibid.). Despite, for

example, widespread rhetorical support for the freedom of expression and the freedom to hold

unpopular views, Stouffer (1955) found, in the era of McCarthyism, that an overwhelming

majority of Americans (nearly two-thirds) supported the firing of employees who were admitted

communists, even if this view held no bearing on their ability to do their job. The compromise of

civil liberties in response to threat goes back further; fears of invasion during World War II led

to the internment of 120,000 Japanese-Americans (Vasi & Strang, 2009). An extensive literature

in political science (Prothro and Grigg 1960; McClosky 1964; McClosky and Brill 1983; Gibson

1988; Sullivan, Piereson, and Marcus 1993; Sniderman et al. 1996; (Vasi & Strang, 2009) finds

similar ambivalence; the “rights for safety” tradeoff called for by NYPD supporters among

residents of high-crime, heavily policed areas, is foreshadowed throughout American history.

Vasi and Strang (2009) review the literature and identify a variety of personal

characteristics associated with support for civil liberties, even in the face of perceived threat.

Community leaders and politically active citizens demonstrate greater support (giving rise to the

“elite theory of democracy”), as do those with greater levels of formal education. Residents of

urban areas also show greater support. Finally, they find some evidence of political influence on

support for civil liberties (though mixed results related to party affiliation), with self-described

liberals showing more support than self-described conservatives.

Vasi and Strang (2009) also note that the tolerance for civil rights impositions is likely to

be tied both to trust in the individuals and institutions implementing these impositions, and the

distribution of burden that might be imposed. Their analysis examined resistance to the USA

PATRIOT Act, which gave state actors extraordinary powers of surveillance, which largely led

to impositions on the civil liberties of immigrants and members of minority groups. They noted

support for entrusting these extraordinary powers to state actors would likely be contingent on

faith that they would be wielded wisely. They also posited that support for these impositions (and

accordingly, resistance to these impositions) revolved in part around principles of inclusivity,

and definitions of “us” and “them”: they anticipated that a portion of the citizenry would be

tolerant of impositions that were borne disproportionately by members of minority groups that

they were not associated with (i.e., “them”, rather than “us”).

This discussion of tradeoffs between indices of self-interest, like safety, and

commitments to symbolic political values such as privacy mirrors a larger long-standing

discussion regarding the role of self-interest in shaping policy attitudes. The symbolic politics

literature (Sears, Lau, Tyler, & Allen, 1980) argues that it is longstanding symbolic attitudes, and

not immediate self-interest concerns, that shape public values. For example, whether a person

might use unemployment insurance does not shape whether they support it. And, whether one

has been a crime victim or feels their neighborhood is not safe has only a minor relationship to

support for law and order policies. Similarly, crime experiences and fears are only weakly related

to support for the death penalty (Tyler & Weber, 1982) and fear of terrorism has little impact

upon attitudes regarding how to police Muslim neighborhoods among either Muslims or non-

Muslims (Huq, Tyler, & Schulhofer, 2011). The degree to which it would be expected that

personal concerns or experiences with crime would shape attitudes about police practices is

unclear.

While the role of symbolic concerns in shaping policy attitudes is well researched (as in

the studies above), the role of self-interest in police-community relationships may manifest in

either of two ways, or a combination thereof. On one hand, people who perceive their

neighborhoods to be dangerous may be willing to sacrifice their civil liberties in the hopes that

some of their perceived danger will be eliminated. However, residents of high-crime, heavily

policed neighborhoods may perceive the police to be a threat that they face beyond the risk of

crime. In such a case, residents at the greatest risk of police contact might indicate less

willingness to sacrifice their rights. The range and strength of self-interest impacts on attitudes

toward the police continues to be issues of contention (Weeden & Kursban, 2017).

Underlying the tensions in the rights-safety tradeoff is the question of whose rights.

Much of the policing controversy of the past two decades, through the recent era of the new

policing, is based on the question of how less powerful groups, those most intensively policed,

can hold accountable the law enforcers whose actions can deprive them of their basic freedoms

of movement and association. Meares (2002) points to the tension between extent of the crime

problem that burdens less powerful and disadvantaged groups, which requires some form of state

intervention and control, and the potential of that intervention to – unintentionally - spill over

into punishment. Punishments in this case range from unnecessary detentions and dignity

incursions to stigma to violence. When disadvantaged groups are asked to sacrifice basic

constitutional protections against such punishment with uncertain returns to their own security

poses risks that shape their views toward the rights tradeoff. One might expect that these groups

would evaluate the rights tradeoff in the context of their shared sense of everyday injustices

(Tyler and Huo, 2002; Fagan, Tyler and Geller, 2014; Fagan, 2017b). While we know that these

experiences erode their views of police legitimacy and weaken cooperation with the police, we

have less understanding of how these concerns are elevated to a broader view of their own

citizenship, the rights they deserve and the rights they believe they are accorded, and their

compliance with the social contract.

These are distant concerns for the less disadvantaged groups with less experience with

the police, but more urgent for those who have greater police contact and who live in the shadow

of crime and violence. The counterfactual to the normative view is that the most intensively

policed citizens may evaluate the tradeoff not in terms of rights, which may be more abstract

concerns, but in more personal terms about their own safety, regardless of the liberty threats to

themselves or their community. Navigating this distinction is one of the challenges to this

research.

C. Current Hypotheses

This paper explores the willingness of urban residents to tolerate impositions on civil

liberties – and specifically, to trade their civil liberties and tolerate police misconduct (hereafter,

to engage in “rights tradeoffs”) in the interest of public safety. Using data collected in a

population-based survey of New York City residents, we test three hypotheses: First, based on

the long American history of compromising civil liberties in the face of perceived threat, we

hypothesize that New Yorkers will report greater are more willing to engage in rights tradeoffs if

they perceive that their neighborhoods are dangerous. Second, as Vasi and Strang (2009) posit

that support for the USA PATRIOT Act would be tied to faith that the government would use its

powers wisely, we hypothesize that New Yorkers will report greater willingness to engage in

rights tradeoffs if they consider the police to be legitimate. Third, expanding on the Vasi and

Strang (2009) concern about “us versus them” distinctions in the distribution of burden, and the

role of self-interest in attitudes toward policing, we hypothesize that New Yorkers will report

greater willingness to engage in rights tradeoffs if they perceive intrusive policing as targeting

“them” rather than “us” – or specifically, if they have little personal experience with the police,

and are at minimal risk of having their own rights violated.

II. METHODOLOGY

A. Data Source

We conducted a phone survey of 960 New York City residents in the summer of 2012,

just after the peak of the NYPD’s Stop and Frisk activity, but before the precipitous decline

observed following the Floyd v. City of New York decision (City of New York, 2016). We used a

stratified sampling plan that divided the city into 146 “neighborhood clusters” and classified

clusters into deciles based on the number of stops the police recorded in 2008 and 2009. Clusters

were randomly selected for sampling within each decile, with a systematic oversample of high-

stop clusters, but including neighborhoods from all 10 deciles of stop activity. Respondents were

selected to be between 25 and 65 years old. This age range was deliberately chosen to be older

than the majority of people stopped by the police (Fagan, 2010; City of New York, 2016), but in

an age range more closely associated with civic engagement, in efforts to assess opinions among

a population that elected leaders would consider to be their core electorate.

Data were collected by phone between June and November 2012. Phone numbers were

chosen from a mix of random digit dial and consumer lists that identified both landline and

cellular numbers more likely to be associated with addresses in the targeted neighborhoods, and

respondents in the desired age range. The survey took approximately 25 minutes, and

respondents received a $15 gift card for their participation. The AAPOR response rate was

between 2 and 3 percent, with a cooperation range between 9 and 12 percent. These rates are

lower than the national averages for 2012 (9% and 14%, respectively – see Pew Research Center

for People and the Press, 2012), reflecting likely reluctance of residents in heavily policed areas

to talk to surveyors about their views of the police (Botero et al., 2013; Richardson et al., 2016;

Schulenberg et al., 2017).

B. Key Measures

1. Support for Rights Tradeoffs

Respondents’ willingness to engage in rights tradeoffs was measured in two ways. First,

respondents were asked explicitly about their views on the tradeoff between public safety and

their civil liberties and given the option to choose one of four responses that best summarized

their views:

1. Respecting my rights is the most important concern.

2. I would sacrifice some of my rights in order to ensure safety from crime.

3. It’s important to balance rights and safety, even if it means compromising a little on each.

4. There is no reason to choose. We can have our rights and the police can keep us safe at

the same time.

We coded a binary indicator of respondents’ willingness to engage in rights tradeoffs,

which was equal to one if respondents selected the second response option (i.e., “I would

sacrifice some of my rights…”), and zero if respondents selected any of the other three options.

A sensitivity analysis includes the third response option (i.e., “It’s important to balance rights

and safety, even if it means compromising a little on each.”) as a second indicator of willingness

to trade off.

We also constructed a scale measuring respondents’ tolerance for officer misconduct

(α=.94), based on eight questions in which respondents could indicate agreement (on 5-point

Likert scales) with statements that “It’s ok for the police to break the law in a [stop/frisk] if [a

variety of circumstances held].” These circumstances, asked as potential justification for

breaking the law in a stop or frisk, included:

1. if it prevents crime.

2. if it helps find criminals.

3. if it helps get rid of guns.

4. if it makes young men think twice about committing crimes.

Responses were coded from 0-4 on each question, with higher values indicating more tolerance

for officer misconduct in the course of stops and frisks.

2. Hypothesized Determinants of Willingness to Engage in Rights Tradeoffs

As noted, we test several hypotheses related to respondent willingness to engage in rights

tradeoffs. First, we hypothesize that respondents will be willing to sacrifice their civil liberties in

the face of perceived threat; we test this empirically by measuring respondents’ perceptions of

their neighborhoods as unsafe. Respondents indicate on a 5-point Likert scale their perceptions

of their neighborhoods, ranging from “very safe” to “very unsafe” with “very unsafe” indicating

the greatest perception of threat. Second, we hypothesize that respondents will be willing to

engage in rights tradeoffs if they see the police as legitimate; we measure perceptions of

legitimacy using an additive scale (α=.74) based on responses to four questions: whether the

NYPD:

1. follows the law in deciding who to stop

2. treats people they stop with respect

3. stops people without a good reason, and

4. treats people differently based on their race.

Each response was coded on a 5-point Likert scale, and coded so that higher values indicate

greater confidence in the legitimacy of the NYPD.

Finally, we hypothesized that respondents will report greater willingness to engage in

rights tradeoffs, and greater tolerance for police misconduct if they are at minimal risk of

actually being forced to trade off their rights or experience police misconduct – specifically, if

their own exposure to the NYPD or aggressive policing practices was limited. We include four

indicators of respondents’ exposures to aggressive police practice:

1. A binary indicator of whether they have been stopped by the police in the past year

2. The number of times they were stopped in the past year

3. Their perception of whether the NYPD stops many young people in their neighborhood

(based on a six-point scale, respondents were coded as 1 if they said the NYPD stopped

“almost all” or “a lot” of young people in their neighborhoods, or 0 if they said the

NYPD stopped “some”, “very few”, “almost none”, or “none” of the young people in

their neighborhood.)

4. A binary indicator of whether they had ever seen someone stopped.

Notably, despite a strong conceptual link between these four indicators of police contact, they are

not highly correlated. The maximum correlation between any two of these indicators is 0.41,

between a respondent having witnessed a stop, and the response that the NYPD stops “almost

all” or “a lot” of the young people in their neighborhood (other pairwise correlations between

these indicators range from 0.15 to 0.39).

3. Potential Confounders

We consider our hypotheses in the context of several demographic and socioeconomic

characteristics commonly associated with exposure to, and attitudes toward the police. We

control for respondent sex (male/female), race and ethnicity (white, black, Hispanic, Asian,

multiracial, or “unknown”), educational attainment (less than high school, high school graduate,

some college or post-high school training, or college graduate), and a binary indicator of whether

the respondent lived in public housing at the time of the survey (Fagan et al., 2012).

4. Analysis Sample

Our analysis is based on two separate analysis samples: We examine respondent

willingness to engage in rights tradeoffs among the 934 respondents who provided an answer to

the relevant survey question. We examine respondent tolerance for police lawbreaking based on

the 885 respondents providing answers to all of the questions in the relevant scale. In each

analysis sample, missing data on predictor values was imputed for 25 imputation datasets, using

the mi impute chained procedure in Stata. Nearly all covariates have fewer than 10% of

observations imputed, with one covariate (whether the NYPD stops “most” or “a lot” of young

people in the neighborhood) imputed on 12% of observations.

C. Modeling Approach

We predict respondent willingness to engage in rights tradeoffs through a series of

logistic and OLS regression models of our two outcomes. For each outcome, we estimate a set of

five models. The first, baseline model estimates willingness to engage in rights tradeoffs as a

function of respondent demographic and socioeconomic characteristics:

(1) Y = β0 + βX + ε

where X = the vector of covariates described above.

We next estimate a model predicting the extent to which respondents express willingness

to trade off their rights as a function of their fear of neighborhood crime. Specifically, Model 2

examines the relationship between willingness to trade and respondents’ perceptions of

neighborhood danger.

(2) Y = β0 + β1DANGER + βX + ε

Our third model tests the hypothesis that respondents will express willingness to engage

in rights tradeoffs if they believe the police will use their powers wisely – or specifically, if they

perceive the NYPD to be legitimate.

(3) Y = β0 + β1LEGIT + βX + ε

Our fourth model tests the hypothesis that respondents will express willingness to engage

in rights tradeoffs if they, personally, are at minimal risk of having their rights violated as a

result. As noted, we operationalize this hypothesis using respondent reports of whether they were

stopped by the police in the past year, the number of times, whether they have witnessed stops of

young people, and whether they think the NYPD stop “a lot” or “almost all” of the youth in their

neighborhoods.

(4) Y = β0 + β1STOPPED + β2NUMSTOPS + β3WITNESS + β4YOUTHSTOPS + βX + ε

Finally, we estimate a model testing our three hypotheses simultaneously. Model 5 tests

the extent to which respondents report willingness to engage in rights tradeoffs based on the

simultaneous influences of fear of crime, trust in police legitimacy, and limitations in their own

exposure to police practices.

(5) Y = β0 + β1DANGER + β2LEGIT + β3STOPPED + β4NUMSTOPS + β5WITNESS +

β6YOUTHSTOPS + βX + ε

It bears noting that because data were collected in a single cross-section, the possibility

exists that long-standing background characteristics, or experiences that predate data collection,

may drive respondents’ perceptions of their neighborhoods and the police, their personal

experience with the police, and their willingness to trade off their civil liberties. We therefore do

not interpret our findings causally, but rather, simply as indicia of which respondents might be

more or less likely to engage in rights tradeoffs.

D. Sensitivity Analyses

We test the sensitivity of our estimated relationships to several of our analytic choices.

First, we consider the possibility that observed relationships (or the lack of significant

relationships) are driven by our choice of correlated predictors; we compute variance inflation

factors (VIFs) of each predictor to identify potentially problematic multicollinearity. We also test

whether our results would substantively change if key predictors were more conceptually

distinguished within our models. We re-estimate Model 4 and Model 5, which test our

hypothesis relating respondents’ thresholds for rights tradeoffs and police misconduct and their

personal exposure to police contact. In one re-estimation, we measure exposure to the police with

a specific focus on recent personal contact (whether the respondent was stopped in the year

leading up to the survey, and how many times), and in another we measure police exposure

based on respondent perceptions of police activity in their neighborhood (whether the respondent

reports that the NYPD stops “most or all” of the young people in their neighborhood, and

whether the respondent reports witnessing a police stop.).

We test the sensitivity of our findings to our coding of one of our outcome measures:

respondents’ willingness to trade off their civil liberties in the interest of public safety. In this

analysis, we construct a more inclusive definition of willingness to trade, in which a respondent

is coded as willing to trade if they either indicate that they “would sacrifice some of [their] rights

to ensure safety from crime”, or that they responded “It’s important to balance rights and safety,

even if it means compromising a little on each”. This more inclusive measure identifies more

than twice as many respondents (33%) as willing to engage in rights tradeoffs, and we assess the

extent to which estimated predictors of willingness to trade are robust to this alternate outcome.

We next assess the extent to which our findings might be driven by our multiple

imputation strategy for dealing with missing data. We re-estimate models of each of our

outcomes using a series of “complete case” samples, including only the respondents who provide

all relevant data for each model. Finally, we assess the extent to which our findings might be

driven by differences between the neighborhoods in which respondents live. To test this we re-

estimate each of our models to include fixed effects for respondents’ neighborhood of residence.

Results from these models will indicate whether the relationships observed across our entire

sample, which may be driven by between-neighborhood differences, are also observed between

respondents living in the same neighborhoods.

III. EMPIRICAL RESULTS

A. Sample Description

A description of the 866 survey respondents included in both our analysis samples is

provided in Table 1. As anticipated given the strategic oversample of high-stop neighborhoods

(which tend to be minority neighborhoods), the sample contains a greater proportion of minority

residents than the city as a whole. Respondent ages span the 25-65 range of our selection criteria,

with an average age of 47.3. Our sample is slightly more educated than the city as a whole, with

88% reporting having completed high school, including 42% who reported completing college or

an advanced degree. 19% of the sample reports living in public housing, which is notable given

the additional discretion that police officers are granted to exercise aggressive police practices in

public housing.

Respondents report substantial exposure to the NYPD: 30% of respondents report that

they were stopped in the year leading up to their interview; those stopped reported an average of

3.3 stops in that time. More than half of respondents (60%) reported having witnessed an NYPD

stop, and 28% reported that the police stopped “a lot” or “almost all” the young people in their

neighborhoods. When asked about their willingness to trade off their rights, 14% reported that

they would sacrifice some of their rights to improve public safety. Tolerance for police

misconduct ranged widely, with nearly 10% reporting strong agreement with all statements that

“it’s okay for the police to break the law…” in stops and frisks under certain circumstances, over

15% reporting strong disagreement under all circumstances, and the remainder of the sample

distributed between the two (mean = 14.3).

B. Model Results

Results from logistics regression models predicting respondent willingness to engage in

rights tradeoffs are presented in Table 2. We see little evidence that respondents will trade off

their rights in the face of perceived neighborhood danger; as shown in Model 2, respondents who

report feeling unsafe in their neighborhoods are no more likely to report willingness to trade off

than those who feel safe. Model 5 indicates a slight increase in willingness to trade off among

those feeling unsafe; however, this difference is only statistically significant at the P<.10 level.

Perceptions of police legitimacy, on the other hand are consistently and significantly related to

respondents’ willingness to trade off their rights; respondents reporting that they see the NYPD

as more legitimate have increased odds of reporting willingness to trade off.

The relationship between respondent exposure to the NYPD and willingness to trade off

is more complex, as shown in Models 4 and 5. Respondents who report being stopped in the past

year are significantly more likely to report willingness to trade off their rights (counter to our

hypothesis). This relationship is diminished with every additional stop the respondent has

experienced; however the declining willingness to trade among respondents experiencing more

stops is only marginally statistically significant (P<.10). Respondents reporting that the NYPD

stops “a lot” or “almost all” of the young people in their neighborhood were also significantly

less likely to report willingness to trade off their rights (in Model 4); however, this relationship

was statistically insignificant in the combined model that also considered perceived NYPD

legitimacy and neighborhood danger (Model 5). Finally, we found no significant difference in

the odds of willingness to trade between those who had and had not witnessed an NYPD stop.

Estimates from OLS models estimating respondents’ threshold for police misconduct are

presented in Table 3. As with the binary measure of willingness to engage in rights tradeoffs, no

empirical support exists for the hypothesis that people who perceive their neighborhoods as

dangerous will be more likely to tolerate police misconduct. Likewise, perceptions of police

legitimacy remain a strong and consistent predictor of respondents’ reported tolerance for the

police breaking the law under certain circumstances. The relationship between respondents’

exposure to the NYPD and tolerance for police lawbreaking is somewhat stronger than the

relationship observed between personal exposure and the binary indicator of willingness to trade

off one’s rights. Respondents reporting that the police stop a lot or almost all of the young people

in their neighborhoods report significantly lower tolerance for misconduct than their

counterparts. However, this relationship is statistically insignificant in the “combined” model.

Examining observed covariates suggests that black respondents are less tolerant of police

misconduct than white respondents, though the association is only statistically significant in a

subset of models predicting tolerance for police misconduct. Notably, Hispanic respondents (and

in some models, Asian respondents) report greater tolerance for police misconduct than white

respondents do. We also observe that more educated respondents (those with more education

than a high school degree) have significantly lower tolerance for police misconduct in stops and

frisks; however, college-educated respondents are more likely to report willingness to trade off

their rights, a difference that is marginally statistically significant (P<.10).

C. Sensitivity Analyses

We tested several alternative measurement and modeling approaches in order to assess

the robustness of our findings. An examination of our OLS and logistic regression models, and

the potential for variance inflation due to multicollinearity suggests that multicollinearity is

unlikely to have a substantial impact on our estimates; Table 4 presents the Variance Inflation

Factors (VIFs) of each predictor variable. The VIF does not exceed 2 for any of our predictors;

each VIF is far below 10, the generally accepted threshold for problem multicollinearity

(Menard, 1995). Nonetheless, we anticipate that the four predictors in Model 4 are at least

conceptually related. The sensitivity analysis examining alternate, more conceptually distinct,

definitions of police exposure is presented in Table 5. This table includes our re-estimations of

Models 4 and 5 to focus, alternately, on respondents’ personal experience with the NYPD

(whether they had been stopped in the past year, and the number of times), and on respondents’

perceptions of what NYPD officers do in their neighborhood (whether they had seen someone

stopped, and whether they thought the NYPD stopped a lot or almost all young people). Table 5

indicates that the findings in Tables 2 and 3 are largely robust to these alternate specifications.

When focusing only on personal contact, a re-estimation of Model 4 indicates that the

relationship between the number of stops respondents have experienced in the past year and both

their willingness to trade off their rights, and their threshold for police misconduct, becomes

statistically significant at P<.05. However, as in our initial estimation, neither of these

relationships remain statistically significant when perceptions of NYPD legitimacy and

neighborhood danger are also considered. In each estimation of Model 5, perceptions of NYPD

legitimacy remain a strong and statistically significant predictor, and perceptions of

neighborhood danger remain unrelated to respondents’ acceptance of rights tradeoffs and police

misconduct.

Results from models examining our more inclusive measure of “rights tradeoffs” are

presented in Table 6. As in the more limited model of rights tradeoffs presented in Table 2,

perceptions of NYPD legitimacy are the most consistently significant predictor of willingness to

engage in our more inclusive measure. Likewise, perceptions of neighborhood danger remain

unrelated to respondents’ willingness to engage in rights tradeoffs. Similarly, respondents

reporting that the NYPD stop lots or most of the young people in their neighborhood are less

willing than their counterparts to engage in rights tradeoffs; however, none of the other measures

of personal exposure to the NYPD are significant predictors of this measure of willingness to

trade.

We next explored whether the results presented in in Tables 2 and 3 were driven by our

imputation of missing data. Complete case models, provided in Tables 7 and 8, indicate that the

estimated relationships (significant and insignificant) are substantively robust to our choice of

missing data strategy. As in the multiple imputation models, our complete case findings indicate

that perceptions of police legitimacy are the strongest and most robust predictor of respondents’

willingness to engage in rights tradeoffs, and threshold for police misconduct. Perceptions of

neighborhood danger are marginally associated with willingness to engage in rights tradeoffs and

tolerate police misconduct (P<.10), but this association is not statistically significant at

traditional levels. The relationship between personal exposure to the police and willingness to

engage in rights tradeoffs remains inconsistent, and does not change substantively from the

multiple imputation findings.

Finally, we tested the extent to which our observed predictors of respondents’ willingness

to engage in rights tradeoffs was sensitive to controls for neighborhood conditions. Tables 9 and

10 present results from a replication of the models in Tables 2 and 3, but including neighborhood

fixed effects. Notably, our main substantive results are unchanged with the inclusion of fixed

effects: respondents who perceive the NYPD as more legitimate are significantly more likely to

report willingness to engage in rights tradeoffs, and show significantly more tolerance for the

police bending the law in the interest of public safety.

IV. CONCLUSIONS

A. Findings and Implications

As often observed throughout American history, despite normative support for the moral

norms and principles of civil rights in the past half century, there are circumstances under which

people are willing to see those rights compromised for themselves and others. We explored these

circumstances using a household survey of New Yorkers near the peak of the NYPD’s Stop,

Question, and Frisk activity. Despite past literature and history suggesting that rights are

compromised in the face of imminent threat, we find very little relationship between

respondents’ perceptions of threat within their neighborhoods, and their willingness to

compromise their civil liberties. Rather, we see that the respondents willing to compromise civil

liberties in the interest of public safety are those who simultaneously perceive the police to

exercise their powers with legitimacy. These contradictory results comport with the extensive

empirical literature on the unevenness of trust in government among Americans, including both

the powerful and the disadvantaged (Cook and Gronke, 2005).

Relatedly, we observe an inverse relationship between perceived police legitimacy and

threshold for police misconduct, suggesting that respondents are most willing to tolerate police

breaking the law in the interest of public safety, only if they think the police exercise their

authority in a lawful, race-neutral, and respectful manner (and presumably, that such violations

will be relatively rare and equitably imposed). Trust in police and tolerance for police

misconduct, then, seems to be contingent on evaluation of the congruence between police values

and those of the citizens they protect and control. Tolerance of police wrong-doing has some

elasticity, which expands or contracts depending on the comportment of police in the course of

their intrusions on liberty. Hidden in this finding is the central importance of police

effectiveness, one of the important components of police legitimacy: police deviation from legal

actions is tolerated so long as it is done without animus or aggression. Conceptually, procedural

features of police interactions loom in the background of the tradeoff decision.

Finally, we find inconsistent evidence that respondents are less likely to report

willingness to trade off their rights, or tolerate police misconduct, if they have greater exposure

to the police; presumably for these respondents (particularly, those stopped multiple times by the

police, and those who report frequent stops of youth in their communities) the imposition created

by the loss of civil liberties is more personally known, the tradeoffs described in the survey

questions are more than hypothetical. Respondents therefore report less tolerant for such

tradeoffs. On the other hand, respondents who report having been stopped, but report only

limited exposure to the police, report greater willingness to trade off their rights.

Our findings suggest that those supporters of invasive policing tactics who justify such

invasion with the reasoning that “people want more” (Kelly, 2012), or that it provides “a

freedom from assault” (MacDonald, 2013) may be disconnected from the personal experiences

of those stopped by the police – particularly those stopped on an ongoing basis. Even staunch

proponents of proactive policing have acknowledged that “[b]eing stopped when you are

innocent is an infuriating, humiliating experience” (MacDonald, 2013), and an extensive

literature on police contact finds that such humiliation is associated with diminished perceptions

of police legitimacy (Fagan & Tyler, 2005; Meares, 2014; Tyler, 2004; Tyler & Fagan, 2008;

Tyler et al., 2014; Tyler & Huo, 2002). It is therefore highly unlikely that the cited calls for more

police intrusion come from those individuals personally affected by intrusive police practices,

and such reasoning must be carefully re-evaluated. This underscores the important of legitimacy

as a mediator of the tradeoff. Research in similar policing contexts suggests that people are

unlikely to voluntarily consent to or approve of such incursions absent a strong measure of moral

duty and obligation, but are less likely to approve of coerced incursions on liberty if rooted

solely on an instrumental concern for safety (Jackson, Bradford, MacQueen, and Hough, 2016).

Beyond the specific issue of policing, these findings support a symbolic perspective on

policies. The results suggest that immediate needs or fears of neighborhood danger do not drive

policy preferences or normative judgments. Rather, people support giving the police authority

when they perceive the police are legitimate, and will tolerate some misconduct under those

conditions. Prior studies on the police show that perceived legitimacy is related to judgments

about the fairness with which the police exercise their authority (Tyler & Fagan, 2008; Tyler,

Fagan & Geller, 2014; ). In other words, legitimacy is not simply a response to instrumental

judgments about the police (they control crime; they deliver desirable outcomes). On the

concrete level of police contact and on the more general level of community wide evaluations of

the police people show that same sensitivity to values that they show when making safety-

privacy tradeoffs.

It is particularly noteworthy that extensive personal contact with the police, and police

presence in one’s neighborhood, is marginally associated with diminished willingness to cede

authority to the police in the name of crime control and safety. Recent discussions of policing

have focused upon the lack of public trust in the police, particularly within minority

communities. Low levels of trust have been associated with a variety of negative outcomes,

including increased criminal conduct, reduced willingness to cooperate with the police, and a

greater likelihood of going outside the law to manage conflicts (Tyler, Fagan & Geller, 2014;

Tyler & Jackson, 2014; Jackson and Gau, 2016). To the extent that extensive personal contact

with the police is associated with diminished community trust, the style of aggressive policing

widely exercised in American cities threatens to undermine support for police policies and

practices among members of the public.

B. Limitations

Our findings to date are limited by features of both our data collection protocol and the

assumptions underlying the analytic models. First, as previously noted, our survey is limited by

low response rates, and it is unclear to whom our findings might generalize. Our sampling plan

was developed to focus on heavily policed neighborhoods, and residents of heavily policed

neighborhoods may be reluctant to discuss local police activity for fear of reprisal (Botero et al.,

2013). Residents who do choose to participate in a household survey, despite these disincentives,

and spend approximately 25 minutes on the phone with an interviewer, are likely to differ from

their neighbors in important ways, observable and unobservable. Those who participated

reported relatively high rates of contact with the police; 30% report having been stopped by the

police in the year before their survey, and 60% report having witnessed an NYPD stop. The

extent to which these experiences and others impact the generalizability of our findings is

unclear.

In addition, our data were collected in a single interview from each respondent. Cross-

sectional data is ill-equipped to assess causal relationships, as observed associations may be

spuriously induced by long-standing characteristics of individuals and their communities, and the

ongoing relationships between individuals, communities, and the police. Moreover, our models

are limited by the correlation between our predictors: the number of stops experienced by

respondents is likely tied to their perceptions of police legitimacy, and personal characteristics

such as their race and residence in public housing. Nonetheless, the general consistency in our

substantive findings across models testing our hypotheses separately, and our combined model,

suggest that this limitation is relatively minor.

C. Conclusion

Both police leaders and politicians more generally often think about public support in

instrumental terms, believing that if they control crime and create perceived safety the public

will not focus on or be concerned about the practices they use to achieve these goals. This paper

supports the argument that this is not the case within the general community, including the most

socially and politically disadvantaged. People are sensitive to police legitimacy which involves

issues of lawfulness and fairness. Legitimacy encourages deference to the police and acceptance

of their discretionary decisions, not the effectiveness of those decisions or the policies they

reflect.

TABLES

Table 1: Sample Description (N=866)

Sample Characteristic

Mean [SE]

or %

Willingness to Trade off Rights Rights Tradeoff Tolerance (binary) 14%

Police misconduct tolerance (scale, 0-32) 14.3

[0.38]

Experience with NYPD Stopped by NYPD, Past Year 30%

Number of stops, past year (among those

stopped) 3.3

[0.4]

Number of stops, past year (including zeros)

1.0

[0.13]

Witnessed NYPD stop of young person 60%

Perceptions of NYPD stops of young people in neighborhood

NYPD stops no kids 22%

NYPD stops almost no kids 8%

NYPD stops very few kids 18%

NYPD stops some kids 15%

NYPD stops a lot of kids 19%

NYPD stops almost all kids 6%

Demographic Background Male 48%

Female 52%

White 30%

Black 39%

Hispanic 24%

Asian 3%

Two or more races 4%

Age 47.3

[0.4]

< HS 12%

HS Only 23%

More than HS, Not College Graduate 23%

College or More 42%

Public Housing Resident 19%

Table 2: Willingness to Trade Off Rights, Odds Ratios

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops (any) 2.118 * 2.224 **

[0.633] [0.669] Past-Year Stops (Number) 0.794 + 0.808 +

[0.100] [0.103] NYPD stops lots or most kids 0.560 * 0.629

[0.158] [0.186] Seen stop take place 1.156 1.263

[0.249] [0.278] Perceived NYPD Legit. 1.100 *** 1.104 ***

[0.027] [0.030] Perceived neighborhood danger 1.021 1.154 +

[0.080] [0.099] Male 1.233 1.237 1.203 1.202 1.156

[0.236] [0.237] [0.233] [0.235] [0.231] Black 0.627 + 0.625 + 0.820 0.648 + 0.796

[0.152] [0.152] [0.209] [0.162] [0.205] Hispanic 0.990 0.984 1.127 0.990 1.052

[0.257] [0.256] [0.298] [0.261] [0.283] Asian 0.566 0.565 0. 489 0.559 0.483

[0.357] [0.357] [0.312] [0.353] [0.310] Two or more Races 0.320 0.319 0.304 0.342 0.321

[0.239] [0.238] [0.229] [0.256] [0.242] Age 1.007 1.007 1.003 1.003 1.002

[0.009] [0.009] [0.009] [0.009] [0.009] < HS 1.502 1.499 1.551 1.540 1.558

[0.534] [0.533] [0.556] [0.554] [0.566] > HS but < college 1.498 1.501 1.591 1.446 1.546

[0.449] [0.450] [0.483] [0.438] [0.474] College + 1.575 + 1.589 + 1.688 + 1.446 1.648 +

[0.429] [0.434] [0.465] [0.398] [0.461] Public Housing 0.892 0.887 0.920 0. 973 0.942

[0.246] [0.245] [0.256] [0.272] [0.267] Constant 0.093 *** 0.090 *** 0.030 *** 0.110 *** 0.023 ***

[0.048] [0.047] [0.018] [0.059] [0.016] N 934 934 934 934 934

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 3: Threshold for Police Misconduct, OLS Coefficients

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops -0.606 -0.033 (any) [0.877] [0.853] Past-Year Stops (Number) -0.193 + -0.130

[0.102] [0.099] NYPD stops lots or most -2.529 ** -1.082 Kids [0.976] [1.001] Seen stop take -0.318 0.362 Place [0.819] [0.806] Perceived NYPD 0.722 *** 0.699 ***

Legitimacy [0.085] [0.095] Perceived neighborhood -0.306 0.395 Danger [0.293] [0.296] Male -0.964 -1.024 -1.332 + -0.815 -1.192 +

[0.717] [0.719] [0.690] [0.723] [0.705] Black -2.820 ** -2.774 ** -0.827 -2.250 * -0.824

[0.906] [0.907] [0.904] [0.918] [0.911] Hispanic 3.161 ** 3.271 ** 4.002 *** 3.489 *** 3.886 ***

[1.035] [1.040] [1.000] [1.033] [1.008] Asian 4.699 * 4.737 * 3.616 + 4.330 * 3.489 +

[2.169] [2.169] [2.089] [2.151] [2.091] Two or more Races 0.001 0.051 -0.402 -0.153 -0.513

[1.982] [1.982] [1.907] [1.967] [1.911] Age 0.105 *** 0.102 ** 0.067 * 0.070 * 0.057 +

[0.032] [0.032] [0.031] [0.033] [0.032] < HS 1.659 1.637 1.725 1.881 1.877

[1.274] [1.274] [1.228] [1.266] [1.230] > HS but < college -3.021 ** -3.095 ** -2.719 ** -2.939 ** -2.602 *

[1.058] [1.060] [1.020] [1.048] [1.023] College + -3.968 *** -4.074 *** -3.583 *** -4.084 *** -3.514 ***

[0.950] [0.956] [0.915] [0.944] [0.925] Public Housing -0.544 -0.425 -0.317 -0.220 -0. 326

[0.956] [0.962] [0.918] [0.951] [0.925] Constant 12.307 *** 12.760 *** 4.592 * 14.803 *** 5.047 *

[1.879] [1.928] [2.023] [1.968] [2.349] N 885 885 885 885 885

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 4: Variance Inflation Factors (VIFs) of Predictor Variables

Variable VIF

College + 1.77

Black 1.65

Hispanic 1.56

> HS but < college 1.54

NYPD stops lots or most kids 1.46

< HS 1.40

Perceived NYPD Legit. 1.36

Past-Year Stops (any) 1.30

Seen stop take place 1.28

Past-Year Stops (number) 1.25

Perceived neighborhood danger 1.19

PH Resident 1.15

Age 1.10

Two or more races 1.09

Asian 1.06

Male 1.06

Mean VIF 1.33

NOTE: VIFs obtained using the mivif command, run after Model 5 predicting tolerance for

police lawbreaking.

Table 5: Sensitivity Analysis: Alternative Specifications of Models Examining Police Exposure

Predicting Binary Measure (Odds Ratios) Predicting Scale Measure (Regression Coefficients)

Personal Contact Perceptions of N’hood Personal Contact Perceptions of N’hood

Predictor Model 4 Model 5 Model 4 Model 5 Model 4 Model 5 Model 4 Model 5

Past-Year Stops

(Any)

2.089

[0.616]

* 2.231

[0.664]

** -1.171

[0.854]

-0.121

[0.837]

Past-Year Stops

(Number)

0.767

[0.097]

* 0.800

[0.102]

+ -0.235

[0.101]

* -0.141

[0.099]

NYPD stops lots

or most kids

0.537

[0.148]

* 0.631

[0.185]

-2.989

[0.948]

** -1.257

[0.985]

Seen stop take

place

1.207

[0.256]

1.342

[0.292]

-0.484

[0.812]

0.326

[0.799]

Perceived NYPD

Legitimacy

1.109

[0.029]

*** 1.104

[0.029]

*** 0.719

[0.090]

*** 0.714

[0.094]

***

Perceived

neighborhood

danger

1.128

[0.094]

1.137

[0.097]

0.343

[0.291]

0.371

[0.296]

Male

1.246

[0.242]

1.186

[0.235]

1.207

[0.232]

1.190

[0.232]

-0.628

[0.722]

-1.142

[0.703]

-1.053

[0.714]

-1.311

[0.693]

+

Black

0.617

[0.151]

* 0.794

[0.203]

0.660

[0.164]

+ 0.812

[0.208]

-2.722

[0.903]

** -0.870

[0.905]

-2.190

[0.919]

* -0.752

[0.910]

Hispanic

0.981

[0.256]

1.071

[0.286]

0.994

[0.260]

1.066

[0.285]

3.308

[1.031]

*** 3.924

[1.003]

*** 3.454

[1.036]

*** 3.890

[1.008]

***

Asian

0.563

[0.356]

0.485

[0.311]

0.551

[0.348]

0.464

[0.299]

4.459

[2.159]

* 3.488

[2.090]

+ 4.446

[2.155]

* 3.528

[2.090]

+

Two or more

0.338

[0.253]

0.322

[0.242]

0.319

[0.239]

0.299

[0.225]

-0.273

[1.975]

-0.540

[1.910]

0.028

[1.968]

-0.449

[1.908]

Age

1.005

[0.009]

1.003

[0.009]

1.004

[0.009]

1.003

[0.009]

0.083

[0.033]

* 0.061

[0.032]

+ 0.082

[0.033]

* 0.063

[0.032]

*

< HS

1.475

[0.528]

1.503

[0.544]

1.548

[0.553]

1.573

[0.566]

1.758

[1.270]

1.825

[1.229]

1.813

[1.267]

1.813

[1.228]

> HS, < college

1.433

[0.433]

1.548

[0.473]

1.494

[0.450]

1.596

[0.487]

-2.959

[1.053]

** -2.617

[1.019]

* -2.972

[1.050]

** -2.616

[1.023]

*

College or more

1.488

[0.409]

1.666

[0.465]

+ 1.510

[0.413]

1.717

[0.478]

+ -3.927

[0.946]

*** -3.467

[0.922]

*** -4.128

[0.945]

*** -3.520

[0.924]

***

PH

0.947

[0.263]

0.935

[0.264]

0.925

[0.256]

0.903

[0.253]

-0.358

[0.953]

-0.352

[0.925]

-0.330

[0.951]

-0.389

[0.924]

Constant

0.099

[0.052]

*** 0.023

[0.015]

0.110

[0.059]

*** 0.024

[0.016]

*** 13.634

[1.916]

*** 4.644

[2.211]

* 14.234

[1.957]

*** 4.591

[2.315]

*

N 934 934 934 934 885 885 885 885

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 6: Sensitivity Analysis: More Inclusive Tradeoff Code (base rate = 33%)

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops (any) 1.090 1.168

[0.199] [0.215] Past-Year Stops (Number) 0.981 0.992

[0.029] [0.028] NYPD stops lots or most kids 0.536 ** 0.646 *

[0.111] [0.142] Seen stop take place 1.249 1.389 +

[0.205] [0.235] Perceived NYPD Legit. 1.097 *** 1.097 ***

[0.020] [0.023] Perceived neighborhood danger 0.935 1.023

[0.057] [0.068] Male 1.200 1.191 1.163 1.167 1.114

[0.171] [0.170] [0.169] [0.170] [0.165] Black 0.773 0.782 1.012 0.811 0.993

[0.138] [0.140] [0.192] [0.149] [0.190] Hispanic 0.974 0.992 1.110 0.974 1.062

[0.198] [0.203] [0.232] [0.201] [0.224] Asian 1.977 + 1.987 1.809 1.939 1.843

[0.794] [0.798] [0.740] [0.783] [0.759] Two or more Races 0.921 0.928 0.893 0.925 0.910

[0.362] [0.366] [0.358] [0.366] [0.366] Age 1.014 1.013 * 1.010 1.010 1.009

[0.007] [0.007] [0.007] [0.007] [0.007] < HS 1.076 1.084 1.103 1.118 1.131

[0.295] [0.298] [0.306] [0.309] [0.316] > HS but < college 1.543 * 1.535 * 1.648 * 1.549 * 1.642 *

[0.334] [0.333] [0.364] [0.338] [0.365] College + 1.870 *** 1.837 ** 2.021 *** 1.807 ** 1.988 ***

[0.364] [0.359] [0.402] [0.354] [0.399] Public Housing 0.790 0.804 0.808 0. 826 0.823

[0.160] [0.164] [0.167] [0.169] [0.172] Constant 0.180 *** 0.196 *** 0.060 *** 0.214 *** 0.055 ***

[0.069] [0.077] [0.027] [0.086] [0.028] N 934 934 934 934 934

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 7: Sensitivity Analysis: Willingness to Trade Off, Odds Ratios, Complete Case Sample

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops (any) 2.647 ** 2.914 **

[0.896] [1.044] Past-Year Stops (Number) 0.738 * 0.761 +

[0.109] [0.114] NYPD stops lots or most kids 0.586 + 0.647

[0.168] [0.204] Seen stop take place 1.075 1.250

[0.237] [0.300] Perceived NYPD Legit. 1.105 *** 1.121 ***

[0.028] [0.032] Perceived neighborhood danger 1.040 1.174 +

[0.081] [0.110] Male 1.246 1.257 1.410 1.223 1.335

[0.240] [0.243] [0.293] [0.248] [0.293] Black 0.592 * 0.589 * 0.750 0.627 + 0.758

[0.145] [0.144] [0.203] [0.162] [0.213] Hispanic 0.967 0.951 1.030 0.928 0.919

[0.253] [0.250] [0.292] [0.253] [0.271] Asian 0.568 0.563 0.554 0.547 0.531

[0.359] [0.356] [0.361] [0.347] [0.349] Two or more Races 0.311 0.308 0.147 + 0.373 0.166 +

[0.232] [0.230] [0.152] [0.281] [0.173] Age 1.008 1.009 1.004 1.002 1.001

[0.009] [0.009] [0.009] [0.009] [0.010] < HS 1.377 1.371 1.201 1.528 1.286

[0.498] [0.496] [0.482] [0.564] [0.530] > HS but < college 1.459 1.483 1.514 1.483 1.490

[0.438] [0.446] [0.487] [0.453] [0.490] College + 1.500 1.521 1.642 + 1.288 1.529

[0.410] [0.418] [0.478] [0.360] [0.457] Public Housing 0.877 0.871 1.072 0. 989 1.112

[0.247] [0.245] [0.317] [0.283] [0.339] Constant 0.093 *** 0.088 *** 0.027 *** 0.128 *** 0.020 ***

[0.048] [0.046] [0.017] [0.070] [0.015] N 916 913 830 870 792

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 8: Sensitivity Analysis: Threshold for Police Misconduct, OLS Coefficients, Complete Case

Sample

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops -0.765 0.182 (any) [0.934] [0.937] Past-Year Stops (Number) -0.186 + -0.118

[0.102] [0.102] NYPD stops lots or most -3.253 *** -2.119 *

Kids [0.976] [1.015] Seen stop take 0.061 0.482 Place [0.840] [0.858] Perceived NYPD 0.728 *** 0.713 ***

Legitimacy [0.086] [0.096] Perceived neighborhood -0.226 0.547 +

Danger [0.295] [0.317] Male -0.910 -0.954 -1.006 -0.759 -0.990 +

[0.722] [0.726] [0.723] [0.750] [0.758] Black -2.689 ** -2.655 ** -0.593 -2.164 * -0.599

[0.916] [0.918] [0.941] [0.953] [0.978] Hispanic 3.421 ** 3.498 ** 4.339 *** 3.829 *** 4.132 ***

[1.045] [0.918] [1.045] [1.067] [1.075] Asian 4.884 * 4.911 * 3.530 + 4.541 * 3.328

[2.170] [2.172] [2.215] [2.154] [2.222] Two or more Races 0.185 0.222 -0.057 -0.173 -0.928

[1.984] [1.986] [2.000] [2.062] [2.077] Age 0.105 *** 0.103 ** 0.065 * 0.057 + 0.040 +

[0.032] [0.032] [0.032] [0.034] [0.034] < HS 1.426 1.419 1.674 1.183 1.381

[1.287] [1.288] [1.294] [1.311] [1.331] > HS but < college -3.157 ** -3.207 ** -2.401 * -2.935 ** -2.362 *

[1.068] [1.072] [1.066] [1.081] [1.091] College + -4.026 *** -4.098 *** -3.433 *** -4.411 *** -3.606 ***

[0.961] [0.966] [0.958] [0.968] [0.980] Public Housing -0.497 -0.414 -0.498 -0.088 -0.295

[0.967] [0.976] [0.964] [0.982] [0.997] Constant 12.165 *** 12.499 *** 4.051 + 15.266 *** 5.239 *

[1.886] [1.937] [2.084] [2.006] [2.444] N 870 869 802 825 764

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 9: Sensitivity Analysis: Willingness to Trade Off Rights, Odds Ratios, Neighborhood

Fixed Effects Models

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops (any) 2.018 * 2.140 *

[0.647] [0.696] Past-Year Stops (Number) 0.789 + 0.805

[0.110] [0.113] NYPD stops lots or most kids 0.521 * 0.575 +

[0.168] [0.186] Seen stop take place 1.063 1.147

[0.247] [0.273] Perceived NYPD Legit. 1.107 *** 1.105 ***

[0.029] [0.031] Perceived neighborhood danger 1.009 1.157

[0.087] [0.109] Male 1.267 1.269 1.236 1.253 1.210

[0.252] [0.253] [0.249] [0.257] [0.251] Black 0.572 + 0.573 + 0.690 0.573 + 0.697

[0.181] [0.181] [0.226] [0.184] [0.231] Hispanic 1.041 1.040 1.112 1.017 1.081

[0.318] [0.318] [0.349] [0.315] [0.342] Asian 0.545 0.545 0.469 0.506 0.455

[0.360] [0.360] [0.314] [0.337] [0.308] Two or more Races 0.276 + 0.275 + 0.249 + 0.302 0.266

[0.215] [0.214] [0.197] [0.237] [0.215] Age 1.013 1.013 1.008 1.007 1.006

[0.009] [0.009] [0.009] [0.010] [0.010] < HS 1.450 1.450 1.598 1.529 1.649

[0.546] [0.546] [0.610] [0.586] [0.639] > HS but < college 1.539 1.541 1.670 1.504 1.653

[0.485] [0.486] [0.535] [0.479] [0.536] College + 1.738 + 1.741 + 1.881 * 1.605 1.799 +

[0.509] [0.511] [0.560] [0.476] [0.544] Public Housing 0.941 0.938 0.976 1.033 0.997

[0.295] [0.296] [0.310] [0.329] [0.324] Constant 0.094 0.093 + 0.026 ** 0.126 0.022 **

[0.117] [0.116] [0.034] [0.161] [0.032] N 881 881 881 881 881

NOTE: +P<.10, *P<.05, **P<.01, ***P<.001

Table 10: Threshold for Police Misconduct, OLS Coefficients, Neighborhood Fixed Effects Models

Model 1 Model 2 Model 3 Model 4 Model 5

Predictor b/se b/se b/se b/se b/se

Past-Year Stops -0.594 -0.063 (any) [0.893] [0.870] Past-Year Stops (Number) -0.193 + -0.130

[0.103] [0.109] NYPD stops lots or most -2.573 *** -1.059 Kids [0.994] [1.016] Seen stop take 0.703 -0.002 Place [0.846] [0.832] Perceived NYPD 0.715 *** 0.676 ***

Legitimacy [0.087] [0.096] Perceived neighborhood -0.402 0.284 Danger [0.307] [0.309] Male -0.891 -0.957 -1.239 + -0.739 -1.108

[0.725] [0.726] [0.698] [0.730] [0.712] Black -2.752 * -2.826 ** -1.332 -2.593 * -1.328

[0.094] [1.095] [1.068] [1.086] [1.074] Hispanic 3.541 ** 3.609 ** 4.001 *** 3.583 ** 3.930 ***

[1.159] [1.160] [1.116] [1.150] [1.119] Asian 4.683 * 4.674 * 3.421 4.034 + 3.245

[2.241] [2.240] [2.162] [2.225] [2.165] Two or more Races -0.490 -0.420 -1.037 -0.772 -1.166

[2.035] [2.034] [1.959] [2.021] [1.967] Age 0.112 *** 0.109 *** 0.072 * 0.074 * 0.060 +

[0.033] [0.033] [0.032] [0.034] [0.033] < HS 1.481 1.436 1.639 1.786 1.818

[1.312] [1.312] [1.265] [1.304] [1.269] > HS but < college -2.959 ** -3.057 ** -2.622 * -2.822 ** -2.517 *

[1.085] [1.087] [1.047] [1.075] [1.051] College + -3.628 *** -3.730 *** -3.265 *** -3.636 *** -3.234 ***

[0.994] [0.996] [0.958] [0.986] [0.964] Public Housing -0.043 0.114 0.086 0.249 0.111

[1.039] [1.045] [0.998] [1.032] [1.007] Constant 9.845 10.209 1.352 11.596 + 2.142

[6.830] [6.383] [6.255] [6.345] [6.346] N 885 885 885 885 885

Note: *P<.05, **P<.01, ***P<.001

REFERENCES

American Civil Liberties Union of Illinois. (2015). Stop and Frisk in Chicago. Retrieved from

Chicago, IL:

Ayres, I., & Borowsky, J. (2008). A Study of Racially Disparate Outcomes in the Los Angeles

Police Department. Retrieved from American Civil Liberties Union,:

Bailey et al. v. City of Philadelphia et al. (2011). Consent Decree. 2010 Civ 05952.

Bayley, David H. (2002) Law enforcement and the rule of law: Is there a tradeoff? Criminology

& Public Policy 2(1): 133-154.

Berg, M. T., Stewart, E. A., Intravia, J., Warren, P. Y., & Simons, R. L. (2016). Cynical Streets:

Neighborhood Social Processes and Perceptions of Criminal Injustice. Criminology,

54(3), 520-547.

Botero, J., Ponce, A., & Shleifer, A. (2013). Education, Complaints, and Accountability. The

Journal of Law and Economics, 56(4), 959-996. doi:https://doi.org/10.1086/674133

Bowers, Josh (2014). "Probable Cause, Constitutional Reasonableness, and the Unrecognized

Point of a Pointless Indignity." Stanford Law Review 66: 987.

Brame, R., Bushway, S. D., Paternoster, R., & Turner, M. G. (2014). Demographic Patterns of

Cumulative Arrest Prevalence by Ages 18 and 23. Crime and Delinquency, 60, 471-486.

Brame, R., Turner, M. G., Paternoster, R., & Bushway, S. D. (2012). Cumulative Prevalence of

Arrest From Ages 8 to 23 in a National Sample. Pediatrics, 129(1), 21-27.

City of New York. (2016). NYPD Stop, Question, and Frisk Report Database. Retrieved from:

http://www.nyc.gov/html/nypd/html/analysis_and_planning/stop_question_and_frisk_rep

ort.shtml

Cook, Timothy E., and Paul Gronke. (2005). "The skeptical American: Revisiting the meanings of trust in government and confidence in institutions." Journal of Politics 67(3): 784-803.

Durlauf, S. N., & Nagin, D. (2011). Imprisonment and crime: Can both be reduced? Criminology

and Public Policy, 10(1), 13-54.

Epp, C. R., Maynard-Moody, S., & Haider-Markel, D. P. (2014). Pulled Over: How Police Stops

Define Race and Citizenship. Chicago, IL: The University of Chicago Press.

Fagan, J. (2010). Expert Testimony. Floyd et al. v. City of New York et al.

Fagan, J. (2016). Terry's Original Sin. University of Chicago Legal Forum, 2016, 43-97.

Fagan, Jeffrey (2017a). "Recent Evidence And Controversies In “The New Policing”." Journal

of Policy Analysis and Management 36(3): 690-700.

Fagan Jeffrey (2017b). "Dignity is the New Legitimacy." In The New Criminal Justice Thinking

(S. Dolovich and A, Natapoff, eds.)

Fagan, J., Braga, A. A., Brunson, R. K., & Pattavina, A. (2016). Stops and Stares: Street Stops,

Surveillance, and Race in the New Policing. Fordham Urban Law Journal, 43(3), 539-

614.

Fagan, J., Conyers, G., & Ayres, I. (2014). No Runs, Few Hits and Many Errors: A Story in Five

Parts about Racial Bias in Stop and Frisk Policing in New York. Paper presented at the

Conference on Empirical Legal Studies, San Francisco, CA.

Fagan, J., & Davies, G. (2000). Street Stops and Broken Windows: Terry, Race, and Disorder in

New York City. Fordham Urban Law Journal, 28, 457-504.

Fagan, J., Davies, G., & Carliss, A. (2012). Race and Selective Enforcement in Public Housing.

Journal of Empirical Legal Studies, 9(4), 697-728.

Fagan, J., Geller, A., Davies, G., & West, V. (2010). Street stops and Broken Windows revisited:

The demography and logic of proactive policing in a safe and changing city. In S. K. Rice

& M. D. White (Eds.), Race, ethnicity, and policing: New and essential readings. New

York, NY: NYU Press.

Fagan, J., & Richman, D. (2017). Understanding recent spikes and longer trends in American

murders. Columbia Law Review, 117(5), 1235-1296.

Fagan, J., & Tyler, T. R. (2005). Legal Socialization of Children and Adolescents. Social Justice

Research, 18(3), 217-242.

Floyd et al. v. City of New York et al. (2013). Liability Decision. 08 Civ 1034 (SAS).

Floyd et al. v. City of New York et al. 959 F. Supp. 2d 540 (2013)

Geller, A. (2018). Policing America's Children: Police Contact and Consequences Among Teens in Fragile

Families. CRCW Working Paper WP18-02-FF. Princeton, NJ: Center for Research on Child

Wellbeing.

Geller, A., & Fagan, J. (2010). Pot as Pretext: Marijuana, Race, and the New Disorder in New

York City Street Policing. Journal of Empirical Legal Studies, 7(4), 591-633.

Geller, A., Fagan, J., Tyler, T. R., & Link, B. (2014). Aggressive Policing and the Mental Health

of Young Urban Men. American Journal of Public Health, 104(12), 2321-2327.

Mary Gutierrez and Shawn Polk v. City of East Chicago and the Housing Authority of the City

East Chicago., Case No. 2:16-cv-00111-JVB-PRC

Hagelskamp, C., Friedman, W., Rizzolo, A., Rinehart, C., & Schleifer, D. (2015). What's at Issue

Here? New York Metro Area Residents on the Problems That Concern Them Most.

Retrieved from New York, NY:

http://www.publicagenda.org/files/PublicAgenda_Whats_At_Issue_Here_2015.pdf

Harcourt, Bernard E. (2004) Unconstitutional police searches and collective responsibility.

Criminology & Public Policy 3(3): 363-378.

Harmon, Rachel A. 2012. "The problem of policing." Michigan Law Review (2012): 761-817.

Herbert, B. (2010). Watching Certain People. New York Times. Retrieved from

http://www.nytimes.com/2010/03/02/opinion/02herbert.html?ref=opinion

Heymann, P. B. (2000). The New Policing. Fordham Urban Law Journal, 28, 407-456.

Hough, M., Jackson, J., Bradford, B., Myhill, A., & Quinton, P. (2010). Procedural justice, trust,

and institutional legitimacy. Policing: a Journal of Policy and Practice, 4, 203-210.

Howell, K. Babe (2009) Broken lives from broken windows: The hidden costs of aggressive

order-maintenance policing. NYU Rev. L. & Soc. Change 33: 271.

Huq, A. Z., Tyler, T. R., & Schulhofer, S. J. (2011). Mechanisms for Eliciting Cooperation in

Counterterrorism Policing: Evidence from the United Kingdom. Journal of Empirical

Legal Studies, 8(4), 728–761. doi:10.1111/j.1740-1461.2011.01239.x

Jackson, J., & Gau, J. M. (2016). Carving up concepts? Differentiating between trust and

legitimacy in public attitudes towards legal authority. In E. Shockley, T. M. S. Neal, L.

PytlikZillig, & B. Bornstein (Eds.), Interdisciplinary Perspectives on Trust: Towards

Theoretical and Methodological Integration (pp. 49-69). New York, NY: Springer.

Jackson, Jonathan and Bradford, Ben, MacQueen, Sarah and Hough, Mike (2016). "Truly Free

Consent? Clarifying the Nature of Police Legitimacy". Available at SSRN:

https://ssrn.com/abstract=2620274

Jain, Eisha (2015). Arrests as regulation. Stanford Law Review 67: 809.

Johnson, Kevin R. "How racial profiling in America became the law of the land: United States v.

Brignoni-Ponce and Whren v. United States and the need for truly rebellious lawyering."

Geo. LJ 98 (2009): 1005.

Justice, B., & Meares, T. L. (2014). How the Criminal Justice System Educates Citizens. The

ANNALS of the American Academy of Political and Social Science, 651(1), 159-177.

Kelling, G. L., & Coles, C. M. (1996). Fixing broken windows : restoring order and reducing

crime in our communities. New York, NY: Free Press.

Kirk, D. S., & Papachristos, A. V. (2011). Cultural Mechanisms and the Persistence of

Neighborhood Violence. American Journal of Sociology, 116(4), 1190-1233.

Krieger, N., Chen, J. T., Waterman, P. D., Kiang, M. V., & Feldman, J. (2015). Police Killings

and Police Deaths Are Public Health Data and Can Be Counted. Plos Medicine, 12(12),

e1001915. doi:10.1371/journal.pmed.1001915

Kubrin, C. E., Messner, S. F., Deanne, G., McGeever, K., & Stucky, T. D. (2010). Proactive

Policing and Robbery Rates Across U.S. Cities. Criminology, 48(1), 57-97.

Langton, L., & Durose, M. R. (2013). Police Behavior during Traffic and Street Stops, 2011.

Retrieved from Washington, DC: http://www.bjs.gov/content/pub/pdf/pbtss11.pdf

MacDonald, H. (2010, June 26, 2010). Fighting Crime Where the Criminals Are. New York

Times, p. A19. Retrieved from https://nyti.ms/2BNVujz

MacDonald, H. (2012). To See Its Value, See How Crime Rose Elsewhere. Room for Debate.

Retrieved from http://www.nytimes.com/roomfordebate/2012/07/17/does-stop-and-frisk-

reduce-crime/to-see-its-value-see-how-crime-rose-elsewhere

MacDonald, John, Jeffrey Fagan, and Amanda Geller (2016). "The effects of local police surges

on crime and arrests in New York City." PLoS one 11(6): e0157223.

Marmar, Charles R., Shannon E. McCaslin, Thomas J. Metzler, Suzanne Best, Daniel S. Weiss,

Jeffery Fagan, Akiva Liberman et al. (2006). "Predictors of posttraumatic stress in police

and other first responders." Annals of the New York Academy of Sciences 1071(1): 1-18.

McClosky, H., & Brill, A. (1983). The Dimensions of Tolerance: What Americans Believe About

Civil Liberties. New York, NY: Russell Sage Foundation.

Meares, T. L. (2002). Signaling, Legitimacy, and Compliance: A Comment on Posner's Law and

Social Norms and Criminal Law Policy. University of Richmond Law Review, 36, 407-

421.

Meares, T. L. (2014). The Law and Social Science of Stop and Frisk. Annual Review of Law and

Social Science, 10, 335-352. doi:10.1146/annurev-lawsocsci-102612-134043

Menard, S. W. (1995). Applied Logistic Regression Analysis: Sage University Series on

Quantitative Applications in the Social Sciences.

Morrow, Weston J., Michael D. White, and Henry F. Fradella. 2017 "After the stop: Exploring

the racial/ethnic disparities in police use of force during Terry stops." Police Quarterly

20.4 (2017): 367-396.

Pew Research Center For the People and the Press. (2012). Assessing the Representativeness of

Public Opinion Surveys. Retrieved from http://www.people-

press.org/2012/05/15/assessing-the-representativeness-of-public-opinion-

surveys/?src=prc-headline

Popkin, Susan J., Olson, Lynn M., Lurigio, Arthur J., Gwiasda, Victoria E., and Carter, Ruth G.

(1995). "Sweeping out drugs and crime: Residents' views of the Chicago Housing

Authority's public housing drug elimination program." Crime & Delinquency 41(1): 73-

99.

Pratt v. Chicago (1994) 848 F.Supp. 792

Rando, J., Broering Derek, Olson, J. E., Marco, C., & Evans, S. B. (2015). Intranasal naloxone

administration by police first responders is associated with decreased opioid overdose

deaths. The American Journal of Emergency Medicine, 33(9), 1201-1204.

Rayman, G. (2010a, Tuesday, May 11, 2010). NYPD Tapes 2: Bed-Stuy street cops ordered:

Turn this place into a ghost town. The Village Voice. Retrieved from

http://www.villagevoice.com/2010-05-11/news/nypd-tapes-part-2-bed-stuy/

Rayman, G. (2010b, Tuesday, June 8, 2010). NYPD Tapes 3: A Detective Comes Forward

About Downgraded Sexual Assaults. The Village Voice. Retrieved from

http://www.villagevoice.com/2010-06-08/news/nypd-tapes-3-detective-comes-forward-

downgrading-rape/

Rayman, G. (2010c, Tuesday, May 4, 2010). The NYPD Tapes: Inside Bed-Stuy's 81st Precinct.

The Village Voice. Retrieved from http://www.villagevoice.com/2010-05-04/news/the-

nypd-tapes-inside-bed-stuy-s-81st-precinct/

Richardson, J. B. J., St. Vil, C., & Cooper, C. (2016). Who Shot Ya? How Emergency

Departments Can Collect Reliable Police Shooting Data. Journal of Urban Health:

Bulletin of the New York Academy of Medicine, 93(1, Suppleent), 8-31.

Rios, V. M. (2011). Punished: Policing the Lives of Black and Latino Boys. New York, NY:

NYU Press.

Rosenfeld, Richard, and Robert Fornango. "The impact of police stops on precinct robbery and

burglary rates in New York City, 2003-2010." Justice Quarterly 31.1 (2014): 96-122.

Sargeant, Elise. (2017) Policing and collective efficacy: the relative importance of police

effectiveness, procedural justice and the obligation to obey police. Policing and society

27(8): 927-940.

Schulenberg, J. L., Chenier, A., Buffone, S., & Wojciechowski, C. (2017). An application of

procedural justice to stakeholder perspectives: examining police legitimacy and public

trust in police complaints systems. Policing and Society, 27(7), 779-796.

Sears, D. O., Lau, R. R., Tyler, T. R., & Allen, H. M., Jr. (1980). Self-Interest vs. Symbolic

Politics in Policy Attitudes and Presidential Voting. The American Political Science

Review, 74(3), 670-684.

Sewell, A. A., & Jefferson, K. A. (2016). Collateral Damage: The Health Effects of Invasive

Police Encounters in New York City. Journal of Urban Health: Bulletin of the New York

Academy of Medicine, 93(1, Supplement), 42-67.

Sewell, A. A., Jefferson, K. A., & Lee, H. (2016). Living under surveillance: Gender,

psychological distress, and stop- question-and-frisk policing in New York City. Social

Science and Medicine, 159(3), 1-13. doi:10.1016/j.socscimed.2016.04.024

Shedd, C. (2015). Unequal City: Race, Schools, and Perceptions of Injustice. New York, NY:

Russell Sage Foundation.

Skogan, W. G., & Frydl, K. (Eds.). (2004). Fairness and effectiveness in policing: the evidence.

Washington, DC: National Academies Press.

Smith, D., & Purtell, R. (2008). Does Stop and Frisk Stop Crime? Lessons from New York City’s

Record Setting Crime Decline. Paper presented at the Association for Public Policy

Analysis and Management, Los Angeles, CA.

Sniderman, P. M., Fletcher, J. F., Russell, P. H., & Tetlock, P. E. (1996). The Clash of Rights:

Liberty, Equality, and Legitimacy in Pluralist Democracy. New Haven, CT: Yale

University Press.

Solove, Daniel J. (2007) I've got nothing to hide and other misunderstandings of privacy. San

Diego Law Review 44: 745.

Soss, J., & Weaver, V. (2017). Police Are Our Government: Politics, Political Science, and the

Policing of Race–Class Subjugated Communities. Annual Review of Political Science,

20, 565-591. doi:https://doi.org/10.1146/annurev-polisci-060415-093825

Stouffer, S. A. (1955). Communism, conformity, and civil liberties; a cross-section of the Nation

speaks its mind. New Brunswick, NJ: Transaction Publishers.

Trinkner, R., Jackson, J., & Tyler, T. (2017). Expanding 'Appropriate' Police Behavior Beyond

Procedural Justice: Bounded Authority & Legal Legitimation SSRN Working Papers,

(2846659).

Tyler, T. R. (2004). Enhancing Police Legitimacy. The ANNALS of the American Academy of

Political and Social Science, 593(1), 84-99.

Tyler, T. R., & Fagan, J. (2008). Legitimacy and Cooperation: Why Do People Help the Police

Fight Crime in Their Communities? Ohio State Journal of Criminal Law, 6(1), 173-229.

Tyler, T. R., Fagan, J., & Geller, A. (2014). Street Stops and Police Legitimacy: Teachable

Moments in Young Urban Men's Legal Socialization. Journal of Empirical Legal Studies,

11(4), 751-785.

Tyler, T. R., & Huo, Y. J. (2002). Trust in the Law: Encouraging Public Cooperation with the

Police and Courts. New York, NY: Russell Sage Foundation.

Tyler, T. R., & Weber, R. (1982). Support for the Death Penalty; Instrumental Response to

Crime, or Symbolic Attitude? Law and Society Review, 17(1), 21-46.

United States. v. Brignoni-Ponce 422 US 873 (1975)

Vasi, I. B., & Strang, D. (2009). Civil Liberty in America: The Diffusion of Municipal Bill of

Rights Resolutions after the Passage of the USA PATRIOT Act. American Journal of

Sociology, 114(6), 1716-1764.

Weisburd, D., & Majmandur, M. (2017). Proactive Policing: Effects on Crime and Communities.

Washington, DC: The National Academies Press. https://doi.org/10.17226/24928

Weisburd, D., Wooditch, A., Weisburd, S., & Yang, S.-M. (2016). Do Stop, Question, and Frisk

Practices Deter Crime? Criminology and Public Policy, 15(1), 31-56.

Whitman, J. Q. (2003). The two western cultures of privacy: Dignity versus liberty. Yale Law

Journal, 113, 1151.

Whren v. United States 517 U.S. 806 (1996)

Williams, V., Fromer, E., Fagbenle, T., & Stein, C. (2012). The Effect of Stop and Frisk in the

Bronx. Radio Rookies. Retrieved from http://www.wnyc.org/story/232447-radio-rookies-

kelly/

Wilson, J. Q., & Kelling, G. L. (1982). Broken Windows. Atlantic Monthly, 249(3), 29-36, 38.

Zimring, F. E. (2011). The City that Became Safe: New York's Lessons for Urban Crime and Its

Control. New York, NY: Oxford University Press.