Content uploaded by Scott Mourtgos
Author content
All content in this area was uploaded by Scott Mourtgos on Oct 04, 2019
Content may be subject to copyright.
Running head: Legal and Community Standards
1
Assessing Public Perceptions of Police Use-of-Force: Legal Reasonableness and
Community Standards
forthcoming in Justice Quarterly
Accepted 10/04/2019
Scott M. Mourtgos & Ian T. Adams
University of Utah
AUTHORS’ NOTE: This article may not exactly replicate the final version published in Justice
Quarterly, as it is not the copy of record.
Scott M. Mourtgos, Department of Political Science, University of Utah;
Ian T. Adams, Department of Political Science, University of Utah.
The authors give special thanks to Dr. Nicholas P. Lovrich for his help and insight in the
preparation of this manuscript.
Correspondence concerning this article should be addressed to Scott M. Mourtgos,
Department of Political Science, University of Utah, Salt Lake City, UT 84112. Email:
scott.mourtgos@utah.edu
LEGAL AND COMMUNITY STANDARDS
2
Abstract
How do public expectations of police use-of-force align with the strict professional and legal
guidelines under which police officers train and operate? This is a largely unexamined but salient
question in the use-of-force literature, and is important given the ongoing public discourse
regarding police use-of-force, community standards, and perceived gaps between the two. This
study focuses on two main research questions: Are substantial portions of the public predisposed
to disapprove of legally reasonable police use-of-force? If so, what are the principal correlates of
those disapproving attitudes? We analyze responses (n=20,781) to General Social Survey (GSS)
questions from 1990 to 2018 entailing police use-of-force scenarios that are prima facie legally
reasonable. We find a substantial proportion of GSS respondents have expressed their
disapproval of legally reasonable, justifiable police uses of force over the entire period, and such
disapproval has increased over time. Causes and policy implications of this misalignment are
discussed.
Keywords: use-of-force; police; citizen perceptions; procedural justice; General Social Survey;
negative binomial regression
LEGAL AND COMMUNITY STANDARDS
3
Introduction
Due to high-profile police use-of-force incidents in the United States over the past several
years, American police have come under intense public and government scrutiny (Shjarback,
Pyrooz, Wolfe, & Decker, 2017). Despite police use-of-force being extremely rare (Eith &
Durose, 2011; Pate & Friddell, 1993), and excessive force comprising just a small proportion of
that, well-publicized use-of-force events have evoked widespread community and professional
concern (Adams, 1999; Alpert & Dunham, 2004; Klinger & Brunson, 2009).
One result of this increased scrutiny is an intensification of effort to understand the
circumstances in which police tend to resort to the use of physical force (Schuck & Rabe-Hemp,
2007; Silver & Pickett, 2015). The resulting research has been presented primarily in a manner
intended to assist practitioners and policy makers in modifying police practices to reduce the
frequency of use-of-force incidents. The advent and now widespread use of Crisis Intervention
Team (CIT) training programs – where officers learn how to respond to situations involving
people with mental illness who are in crisis – and the incorporation of such training in police
academy curricula reflect this interest in reducing the likelihood of police resorting to the use of
physical force in police/citizen encounters. Yet, the police remain the state’s primary agent of
coercive force, and the police are often called because someone present at the scene believes that
coercive force may be necessary to resolve a problematic situation (Herbert, 2006). Officers are
trained and expected to use a reasonable amount of physical force required to arrest individuals
in accord with the prevailing circumstances. Moreover, police officers are permitted to protect
themselves and others from harm with the use of reasonable force when necessary (Thompson II,
2015).
LEGAL AND COMMUNITY STANDARDS
4
A largely unexamined but salient question in the use-of-force literature is this: What if
public expectations regarding police use of reasonable force are ill-informed? Is there evidence
indicating that some portion of the public entertains unreasonable expectations regarding police
use-of-force? More specifically, does a noteworthy portion of the public find fault with police
use-of-force actions that are legally reasonable, given the strict professional standards, training,
and legal guidelines under which police officers operate? This study explores the gap between
the legally reasonable standards that officers operate within, and public expectations of police
use-of-force.
An examination of General Social Survey (GSS) data regarding police use-of-force is
presented to delve into this largely unexplored question. One particularly relevant question
regularly included in the GSS over several decades is this: “Are there any situations you can
imagine in which you would approve of a policeman striking an adult male citizen?” It is
unreasonable to expect that a police officer would never find it necessary to strike an adult male
citizen in the course of their work. Arresting criminal suspects can become a violent undertaking
in a range of circumstances such as breaking up bar fights, responding to domestic violence calls,
handling drunk and disorderly calls, making drug arrests, defending themselves from assault, and
a variety of other commonly encountered situations. Indeed, the FBI reports that in 2016 alone
officers were assaulted at a rate of 9.8 per 100 sworn officers (FBI, 2017). Injuries requiring
emergency room visits among US police officers occur at a yearly rate of 635 per 10,000
officers, compared to a non-officer rate of 213 per 10,000. Assaults are the single most
significant driver, responsible for 35% of those officer injuries (Tiesman, Gwilliam, Konda,
Rojek, & Marsh, 2018).
LEGAL AND COMMUNITY STANDARDS
5
Despite this stark reality, an astonishing 34% of respondents who provided an answer to
the above question in the 2018 iteration of the GSS answered “no.” That is, they indicated they
could not imagine any situation in which they would approve of a policeman striking an adult
male citizen. What is particularly noteworthy is the fact that there is a clear upward trend in the
“no” response to this question since it was first asked in 1973, as illustrated in Figure 1.
Figure 1
Figure 1 depicts respondents’ declining support over time for a general question
regarding police use-of-force, but the GSS includes questions with more context as well. For
instance, are officers ever justified in striking an escaping subject, or striking a citizen who is
attacking the officer? These questions and others are examined in much more detail later, but for
now, we immediately see that legally reasonable hypothetical scenarios of police use-of-force
have seen substantially weakened public support over the decades of GSS administration.
LEGAL AND COMMUNITY STANDARDS
6
Figure 2
This cursory examination of GSS longitudinal data indicates that a sizeable portion of the
American public is wont to disapprove of police use-of-force even when such action may be
legally reasonable and in full accord with police professional training. Of course, while this
simple approve/disapprove format may overstate or understate respondents’ actual views on this
matter (Cullen, Fisher, & Applegate, 2000), the importance of this observation is substantial.
Mismatched viewpoints regarding use-of-force incidents can lead to serious tensions
between and among the police, their departments, and the communities they serve (Alpert &
Dunham, 2004; Alpert & Smith, 1994). This discrepancy is deeply concerning during the current
times of heightened racial and ethnic tensions. Mutual trust between the police and the public is
essential for maintaining ongoing collaboration and cooperation between law enforcement and
the public (Moon & Zager, 2007). While much of the police-community trust literature focuses
on public trust in the police, understanding what affects police trust in the public is just as vital
LEGAL AND COMMUNITY STANDARDS
7
since the police-community relationship is reciprocal and interactive (Bottoms & Tankebe, 2012;
Mourtgos, Mayer, Wise, & O’Rourke, 2019; Sargeant, Murphy, & Madon, 2018).
Public perceptions that diverge substantially from legally reasonable standards applied to
police conduct have potentially adverse consequences for both police legitimacy and the
promotion of procedural justice. The President’s Task Force on 21st Century Policing (2015)
recognized legitimacy as one of the key pillars of effective policing, and viewed procedural
justice as being central to creating and maintaining that legitimacy. Police legitimacy is
undoubtedly crucial; people who believe that the police are legitimate are more likely to follow
the law(s), comply with police, cooperate with the police in prosecuting criminal cases, and
accept police decisions (Worden & McLean, 2014). It is commonly recognized that the “key
antecedent of legitimacy is the fairness of the procedures used by the police [i.e., procedural
justice]” (Sunshine & Tyler, 2003, p. 513), and “[w]hen the actions [of police] seem to be
arbitrary, racially disproportionate, or unjustified by the actual situation, they are more likely to
be perceived by the public (and by the Department of Justice) as excessive” (Owens, Weisburd,
Amendola, & Alpert, 2018, p. 44). In other words, evaluations of police use-of-force are often
examined from the viewpoint of a prevailing community standard, not that of a purely legal
viewpoint. What the literature on police legitimacy and procedural justice has left mostly
unexamined, unfortunately, is whether those community standards are constructed in a manner
that aligns to a marked degree with the professional training, standards, and legal guidelines
under which police officers operate.
Complicating the analysis is recent work that recognizes that procedural justice applies
not just to the actions of officers, but the actions of citizens as well. When citizens treat officers
in procedurally just ways, officers are more likely to perceive members of the public as less
LEGAL AND COMMUNITY STANDARDS
8
threatening and are more willing to be mindful of procedural justice themselves (Pickett & Nix,
2019). When police officers believe they are employing use-of-force in a procedurally just
manner, but endure criticism for their justified actions, they may come to believe they are being
treated in a procedurally unjust manner. Where a sizeable gap between legal standards and
prevailing community standards exists, there is a clear danger of a vicious cycle phenomenon
arising, causing police/community mutual distrust to impair public safety at the detriment of the
police and the public alike (Mourtgos et al., 2019).
If community standards and the professional and legal standards under which police
operate are in misalignment, police officers are placed in an untenable position. To which
standard should police departments and police officers hold themselves? To complicate the
matter of community standards even further, the processes in place for the review of police use-
of-force actions are not uniform across local jurisdictions or demographic and cultural sub-
populations. That is, the expectations of different audiences (i.e., racial/ethnic and partisan
political groups) may conflict (Alpert & Smith, 1994; Roche & Roux, 2017), potentially
producing an inconsistent community standard to abide by for police officers and agencies.
The preceding discussion suggests why the concept of reasonable force is so difficult to
define in practice (Adams, 1999). Judgments about reasonable force are inescapably subjective
(Alpert & Smith, 1994), and are likely influenced by personal history, global beliefs about
police, and other psychological and sociological factors unrelated to the facts of the case. One of
the main elements of police legitimacy is the judgment that police policies and actions comport
with established legal rules (Beetham, 1991; Bottoms & Tankebe, 2012; Gilley, 2006; Nivette &
Eisner, 2013; Tankebe, 2013). It is therefore essential to ascertain what is considered reasonable
LEGAL AND COMMUNITY STANDARDS
9
under the law and common law precedent in adjudicated cases in dispute, not necessarily what
may be deemed reasonable in the court of public opinion.
Of course, examining how members of the public assess legally reasonable use-of-force
actions does not resolve the possible community standard versus legal standard tension described
above. In an exploratory study such as this, however, the legal standard is the appropriate place
to begin for an examination of lay citizen evaluations of police use-of-force. It is the standard
which police are professionally trained to uphold, and by which prosecutors, juries, and courts of
law ultimately judge police actions. If significant discrepancies exist between what is legally
reasonable and what citizens believe is normatively reasonable, further research is warranted to
understand more fully, and address more effectively, this gap in expectations. We return to this
possible mismatch between legal reasonableness and community standards in the discussion
section of this article. First, though, it is useful to examine the concept of legally reasonable
force, and how it interacts with more subjective evaluations of reasonableness.
Reviewing Legal Standards of Reasonable Use of Force by Police
The most important legal consideration for police officers charged with excessive use-of-
force is the U.S. Supreme Court case Graham v. Connor (1989). In Graham, the Court laid out a
partial list of the factors to be considered in determining the legal reasonableness of police use-
of-force situations. The court laid out a three-part, non-exclusive test for assessing use-of-force
reasonableness, “… including the severity of the crime at issue, whether the suspect poses an
immediate threat to the safety of the officers or others, and whether he is actively resisting arrest
or attempting to evade arrest by flight.” Further, the “reasonableness of a particular use of force
must be judged from the perspective of a reasonable officer on the scene, rather than with the
20/20 vision of hindsight…The calculus of reasonableness must embody allowance for the fact
LEGAL AND COMMUNITY STANDARDS
10
that police officers are often forced to make split-second judgments – in circumstances that are
tense, uncertain, and rapidly evolving – about the amount of force that is necessary in a particular
situation” (Graham v Connor, 1989, p. 396-397).
In other words, under the Graham doctrine, an officer’s actions are judged in comparison
to other similarly trained and experienced officers in the same situation, knowing what the actual
officer knew at the time – not what circumstances the officer, or the public, learned after the fact.
The perspective of a reasonable officer at the scene is a restatement of the legal doctrine that
members of a profession must act with the skill and judgment commonly possessed by members
of that profession (Engel & Smith, 2009).
While police are trained to endeavor to exhaust available alternatives to the use of
physical force to the extent possible, under the Graham doctrine officers are not required to use
less intrusive force if the use-of-force in question would be deemed reasonable under the Fourth
Amendment (Thompson II, 2015). The Ninth Circuit Court of Appeals has held that in many
situations “requiring officers to find and choose the least intrusive alternative would require them
to exercise superhuman judgment” (Scott v. Henrich, 1994, p. 915).
The Graham doctrine provides a professional and legal understanding of reasonableness
viewed as an objective evaluation. Under this understanding, the putatively objective evaluation
is “…based solely on the objective circumstances of the event and the conclusion that would be
drawn by any ‘reasonable officer at the scene’” (International Associations of Chiefs of Police,
2006, p.2). Graham does not entirely remove the subjectivity of human evaluation, but instead
provides a general framework for the legal system to evaluate whether an officer’s actions are
“objectively reasonable” based on the circumstance confronting an officer, not the officer’s
subjective intent or motive. More precisely, the legal question is: Given the totality of the
LEGAL AND COMMUNITY STANDARDS
11
objective circumstances presented to the officer at the scene, would a reasonable officer take the
same actions (Shane & Swenson, 2019)?
The formulation of the “reasonable officer” standard, however, has remained problematic
on several fronts from Graham’s inception. First, there is the definitional ambiguity of what a
“reasonable officer” is. This ambiguity created by the Graham decision was intentional, as the
Court acknowledged “reasonableness…is not capable of precise definition or mechanical
application.” Different officers at the same scene may view the dynamics of a use-of-force
encounter differently, hence responding with different actions (e.g., the 2014 shooting of Laquan
McDonald in Chicago; Alpert & Smith, 1994). Comparably trained and competent officers may
respond with different actions based on normal physiological responses that create perceptual
distortions when persons act under extreme stress (Klinger & Brunson, 2009). Further, the
reasonable officer standard may not permit the effective remediation of systemic problems of
persistent use of excessive force since the standard is based on what actions most officers would
take in similar circumstances, including those who engage in excessive use of force.
Second, the “reasonable officer” standard is not being judged by an officer’s peers and
professional police administrators when a claim of excessive force is adjudicated in a jury trial,
where ultimately non-police citizens determine what a reasonable officer should have done in the
case at hand. Clearly, police officers and members of the public do not share a common pool of
experience (Klinger & Brunson, 2009). Accordingly, an officer’s actions will be viewed through
the lens of a “civilian reasonable person” standard which may differ significantly from an
experienced officer’s standard, and “[i]t is likely that many reasonable and even highly-skilled
officers would respond similarly in a given factual scenario that a jury has determined to be
unreasonable conduct” (Alpert & Smith, 1994, p. 491). This discrepancy in experience between
LEGAL AND COMMUNITY STANDARDS
12
police officers and members of the public often leads to the employment of use-of-force subject
matter experts in court cases adjudicating instances of police use-of-force. Attorneys defending
officers often feel the need to use these experts in order to explain officers’ actions.
Police officers commonly believe that the public does not understand the realities of their
job, including instances where force is used (Bartels & Silverman, 2005). A common sentiment
expressed by officers is that “the public is generally naïve about police work…Members of the
public are basically unsupportive and unreasonably demanding. They all seem to think they
know our job better than we do” (Sparrow, Moore, & Kennedy, 1990, p. 51, emphasis supplied).
Regarding the use-of-force issue specifically, recent research has found that police
officers tend to express doubts regarding citizens’ ability to understand and make competent
judgments about police use-of-force. This perception on the part of police officers is a significant
driving factor behind levels of police trust in the public (Mourtgos et al., 2019). A commonly
shared view police officers have of the public’s use-of-force expectations is, “[T]here are people
sometimes that misunderstand us, and look at any use of force as unreasonable. That’s not a
reasonable expectation to have of a police force” (Wieber, 2017, para. 17). Bolstering this point,
a recent survey by the Pew Research Center of a representative sample of U.S. police officers
lends credence to this presumption of doubt by police of the public’s ability to make informed
judgments on police use-of-force. The Pew survey found that only 14% of police officers taking
part in their national survey believed that citizens tend to have an adequate understanding of the
risks and challenges that police officers face (Morin, Parker, Stepler, & Mercer, 2017).
In essence, the Graham framework has resulted in a standard described by Alpert and
Smith (1994) as “subjective objectivity.” At the institutional level, the legal system evaluates the
reasonableness of police officers’ use-of-force based on the objective circumstances presented to
LEGAL AND COMMUNITY STANDARDS
13
a given officer at a given scene. However, at the individual level, the incident is evaluated
subjectively by various stakeholders (i.e., judge, jury, police officer, member of the public).
These subjective opinions and standards may not (and often do not) align, and result in conflict
and tension in the ongoing relationship between the police and the public they serve. Graham v.
Connor (1989) is thirty years old, and there are recent calls to restructure the ‘reasonable’
standard with a ‘necessary’ one (Adler, 2019). A move to restructure the legal basis for force is
an unsurprising development given the trend reported here of decreasing support for police use-
of-force overall during the past several decades.
This tension provokes our interest in the research presented here. Examining how citizens
determine legally reasonable use-of-force does not resolve the issue of “subjective objectivity.”
An objective definition of reasonable force continues to remain unreachable, leaving only limited
guidance for police officers and police administrators (Alpert & Smith, 1994), which suggests
that the current legal standard for reasonable police use-of-force is the correct place to start an
examination of this topic.
There is not yet sufficient literature on the mismatched expectations of citizens and the
police regarding use-of-force standards. Previous scholarship examining attitudes toward police
generally, and a robust literature examining public sentiments on police use-of-force issues,
provide a starting point for generating hypotheses for what factors may influence citizen attitudes
toward use-of-force standards. We briefly review race and minority ethnicity, political leanings,
and level of formal educational attainment as possible variables that have an impact on citizen
expectations regarding use-of-force standards, and provide hypotheses for each.
Race and Minority Ethnicity
LEGAL AND COMMUNITY STANDARDS
14
Race is one of the most common correlates studied in the area of police use-of-force
policy preferences and incident assessments. Research regarding race and police use-of-force has
universally shown that African-Americans and other people of minority racial/ethnic status tend
to have a more negative view of the police and police use-of-force than their Non-Hispanic white
counterparts (Arthur & Case, 1994; Brown & Benedict, 2002; Halim & Stiles, 2001; Hurst,
Frank, & Browning, 2000; Roche & Roux, 2017; Thompson & Lee, 2004; Tuch & Weitzer,
1997). While there are numerous plausible reasons for these well-documented negative views, it
is generally argued that they stem principally from historical police mistreatment of racial
minorities (Hurwitz & Peffley, 2005; Johnson & Kuhns, 2009), dating back to slave patrols in
pre-Civil War America and continued biased policing up to the present time (Hadden, 2001). As
such, we hypothesize:
Hypothesis 1: Black and other minority race respondents are more likely than white
respondents to disapprove of legally reasonable police use of force.
Political Leanings
The effect of political predispositions on citizen perceptions of police use-of-force
incidents has been far less often studied than race and ethnic minority status. However, it has
been shown that individuals with conservative views are more likely to hold favorable opinions
of the police than those with liberal political views (Tyler, Boeckmann, Smith, & Huo, 1997).
Several political psychologists have suggested that these ideological divides emerged due to the
institution of law enforcement embodying such conservative values as law and order, respect for
authority, and the protection of private property and proprietary information. In contrast, liberal
values tend to reflect egalitarianism, right to protest and dissent from existing norms and rules,
LEGAL AND COMMUNITY STANDARDS
15
and transparency in government and business activities and archival information (Braithwaite,
1998; Davis & Silver, 2004; Graham, Haidt, & Nosek, 2009).
Recent work by Patil (2018) shows that individuals who self-identify as politically liberal
are more likely than their self-identifying conservative counterparts to favor punishing officers
who have committed a so-called false-positive error (i.e., shooting an unarmed suspect) in a
factually ambiguous situation. Braga, Winship, Tyler, Fagan, and Meares (2014) found that the
more strongly individuals identified as politically conservative, the more approving their
assessment of police use-of-force as depicted in real-life police videos. Fine, Rowan, and
Simmons (2019) found that political orientation affects youths’ perceptions of police as well;
Democratic youth hold more negative perceptions of police than do Republican youth. Although
the evidence is relatively scant, in line with the above observations we hypothesize:
Hypothesis 2: Respondents with liberal political predispositions are more likely,
and those with conservative leanings less likely, to disapprove of legally reasonable
police use-of-force when compared to those with moderate or middle-of-the-road
political inclinations.
Formal Education
Lower class status citizens are the most frequent targets of police abuse (Alpert & Smith,
1994). Level of formal education is widely used across the social science disciplines as a proxy
measure of socioeconomic status. However, the empirical evidence on the effect of education is
mixed in the case of perceptions of the police and police use-of-force. While some evidence
indicates a positive relationship between education and perceptions of police and police use-of-
force (Arthur & Case, 1994; Halim & Stiles, 2001; Silver & Pickett, 2015), other findings
produce a more complex picture. For example, there may be interactions between race and
LEGAL AND COMMUNITY STANDARDS
16
educational levels, as suggested by evidence that more highly educated black persons perceive
more injustice in police actions than their less educated counterparts (Weitzer & Tuch, 1999;
Weitzer & Tuch, 2002).
Adding further to the diversity of prior findings, some studies suggest that education has
no appreciable effect (e.g., Johnson & Kuhns, 2009), while Braga et al. (2014) find formal
educational attainment was the sole control variable with a statistically significant relationship
across respondents’ assessments of police use-of-force scenarios. In the Braga et al. study,
respondents with a college degree or higher were significantly more likely to register more
disapproval of police use-of-force than those with lower education levels.
A review of the relationship between formal educational attainment and perceptions of
police use-of-force shows the empirical record is mixed, pointing to the need for more in-depth
investigation. Given the scope of this study, the data available, and these blended results as a
backdrop, it is difficult to provide an a priori hypothesis. However, on balance, the literature
appears to indicate a positive relationship between higher levels of education and evaluations of
police use-of-force (Arthur & Case, 1994; Halim & Stiles, 2001; Silver & Pickett, 2015). As
such, we posit the following hypothesis:
Hypothesis 3: Respondents with higher levels of formal education will be less likely
than respondents with lower levels of formal education to disapprove of legally
reasonable police use of force.
Given the above guidance, we turn to a presentation of the research questions addressed
in this article and explore several decades of survey data regarding public views on police use-
of-force issues. We document the patterns uncovered in this exploration, including how they
comport with the hypotheses set forth above. We then suggest some reasons for the discrepancies
LEGAL AND COMMUNITY STANDARDS
17
found between citizen expectations and legal standards for police conduct. We close with a
discussion of possible policy actions to be considered to reduce the gap in expectations, along
with future research directions to be considered in deepening our knowledge in this vital area of
scholarship.
The Present Study
This study focuses on two main research questions: Are substantial portions of the public
predisposed to judge legally reasonable police use-of-force disapprovingly? If so, what are the
principal determinants of those disapproving attitudes?
To address these questions, we conducted a negative binomial regression analysis of
General Social Survey data for the period 1990 – 2018. The study uses 1990 as the initial year
for analysis because it is one year after the Graham v. Connor (1989) precedent-setting decision.
Beginning our analysis with the year 1990 allows for a one-year lag for respondents to become
aware of the still-intact-today Graham rules that govern police use-of-force. We do not expect
members of the general public, or even individual police officers, to be familiar with the
intricacies of the Graham doctrine, but the ruling is a convenient demarcation of the modern U.S.
police use-of-force era. Police officers of today are trained in, and ultimately judged by, the
standards specified in the Graham doctrine. Thus, limiting the dataset to this period places the
analysis of both officer expectations and public perceptions on an equal temporal footing.
Controls
Several controls are used to account for possible confounding factors. First, because prior
research has consistently shown that men tend to approve of police use-of-force more often than
women (Arthur & Case, 1994; Barkan & Cohn, 1998; Thompson & Lee, 2004), we control for
the GSS respondents’ sex. Second, the majority of police-related perception studies investigating
LEGAL AND COMMUNITY STANDARDS
18
respondent age as a variable find that younger persons tend to view the police less favorably and
with greater suspicion than older persons (e.g., Brown & Coulter, 1983; Cao et al., 1996;
Chermak, McGarrell, & Weiss, 2001; Murphy & Worrall, 1999; Worrall, 1999). Given these
findings from prior studies, we control for respondent age. Third, since previous research has
shown that those in lower-income categories tend to have more negative views of the police than
individuals in the middle- or higher-income categories (Brown & Coulter, 1983; Cao, Frank, &
Cullen, 1996; Parker, Onyekwuluje, & Murty, 1995), we control for income. Fourth, while there
is a limited amount of evidence regarding differing urban and rural views of the police, police-
public tensions over use-of-force tend to be particularly salient in urban high-crime areas where
police use-of-force is relatively frequent and often forceful (Geller & Scott, 1992; Klinger &
Brunson, 1992). This place-based disparity may suggest that rural residents might view police
use-of-force more favorably than their urban counterparts (Huang & Vaughn, 1996; Worrall,
1999). Given this possibility, we control for the place of residence descriptor featured in the GSS
data. Finally, we include the year of the survey (i.e., 1990 – 2018) as a final control variable.
Data and Methods
The Survey
This study utilizes data gathered by the National Opinion Researcher Center’s General
Social Survey (GSS). The GSS provides a nationally representative sample survey of U.S.
households conducted on a biennial basis. The GSS uses a full-probability sample, and selection
occurs via a multi-stage sampling of households. The GSS uses a split-sample design and
consists of three different ballots. The split-sample design allows for the introduction of new
topics during each survey period, while maintaining a consistent subset of questions across all
LEGAL AND COMMUNITY STANDARDS
19
surveys collected in a given year. Selection of ballots for respondents is random, so there is no
systematic bias affecting which survey version a particular respondent receives.
We constrain results to the post-Graham years of 1990 through 2018. The GSS total
response count for that period is (n= 39,921), but due to the split-sample design of the GSS, the
total sample size for this study is limited to respondents presented with the five police use-of-
force questions at the heart of this study (n = 20,781). Regression analysis results in dropping
809 observations, but the overall sample is compellingly large (n = 19,972), such that effort to
recover the dropped observations is not undertaken.
Measures of Assessments of the Reasonability of Police Use-of-Force
Legally Reasonable Use-of-force. Each respondent in the sample was asked the following
five questions regarding police use-of-force:
1. POLHITOK: Are there any situations you can imagine in which you would approve of a
policeman striking an adult male citizen?
2. POLABUSE: Would you approve of a policeman striking a citizen who had said vulgar
and obscene things to the policeman?
3. POLMURDR: Would you approve of a policeman striking a citizen who was being
questioned as a suspect in a murder case?
4. POLESCAP: Would you approve of a policeman striking a citizen who was attempting to
escape from custody?
5. POLATTAK: Would you approve of a policeman striking a citizen who was attacking the
policeman with his fists?
The responses are coded as yes = 0 and no = 1, with the yes response indicating that the
respondent approved of the use-of-force action in question. Past studies utilizing the five GSS
questions regarding police use-of-force have applied varying methods to the analysis of this
block of survey items. Some researchers have examined each of the five questions individually
(Halim & Stiles, 2001; Thompson & Lee, 2004), while others have created a scale based on
adding responses to all or some of the questions (Carter & Corra, 2016; Wilson & Dunham,
LEGAL AND COMMUNITY STANDARDS
20
2001). Yet others have suggested that the contexts provided by the different questions may lead
to different evaluations of the appropriateness of the use-of-force. In reflection of this specific
presumption, they separated the several questions accordingly (Barkan & Cohn, 1998; Johnson
& Kuhns, 2009).
Following Barkan and Cohn (1998) and Johnson and Kuhns (2009), we apply factor
analysis to all five questions. While Barkan and Cohn (1998) and Johnson and Kuhns (2009)
only factor analyzed POLABUSE, POLMURDR, POLESCAP, and POLATTAK, we included
the POLHITOK item as well. While the other four questions are decidedly context-specific, and
POLHITOK is a general specification of normative conduct, its inclusion provides valuable
insight into the public’s judgment of police use-of-force within our conceptualization.
A brief discussion about which questions we expected to load on a ‘legally reasonable’
factor is necessary. We hypothesized that POLHITOK, POLESCAP, and POLATTAK would all
load together on one ‘legally reasonable’ factor. First, as discussed previously, it seems legally
and practically unreasonable to expect a police officer to never strike anyone in the course of
carrying out their duties (POLHITOK). Second, if someone is trying to escape custody, it would
seem legally and practically reasonable to expect that a police officer may have to strike that
individual to keep them from escaping and posing a danger to the community (POLESCAP)
(Alpert & Smith, 1994). Finally, it would seem highly unreasonable, both legally and morally,
for an officer not to strike someone who was attacking them with their fists; there is a well-
established and broadly recognized expectation of the right to self-defense (POLATTAK). On
the other hand, it would be legally unreasonable within the discussion of ‘reasonableness’ above
for a police officer to strike someone who was speaking rudely to them (POLABUSE) or being
interviewed as a murder suspect (POLMURDR).
LEGAL AND COMMUNITY STANDARDS
21
It should not be assumed that the specific questions featured in the GSS generalize to
attitudes for all forms of police use-of-force. Striking a citizen is but one type of force, and
attitudes on striking actions may not align with attitudes toward police use of lethal force, or of
less-lethal weapons such as batons and tasers. However, because the GSS datasets feature only
responses pertaining to the act of striking a citizen, we are limited to an examination of this
specific type of force. This limitation in the GSS data should be kept in mind.
We test our hypotheses regarding item loadings with factor analysis. Results of the factor
analysis are presented in Table 1 below. To avoid potential bias due to the dichotomous nature of
all five items, the tetrachoric correlation method was the most appropriate technique for
performing factor analysis in this study (Mislevy, 1986; Muthen & Hofacker, 1988). Based on
the tetrachoric correlation matrix of the five items, maximum-likelihood factor analysis was
conducted with oblique rotation (direct oblimin). The Kaiser-Meyer-Olkin measure verified the
sample adequacy of the analysis (KMO = .6). All KMO values for individual items ranged
between the minimally acceptable limit of .5 to the mediocre level of .7 (Field, 2013). Though
short of the “meritorious” (>0.8) or “marvelous” (>0.9) levels originally defined by Kaiser
(1974, p. 35), more recent methodological advice suggests the KMO values found here range
between mediocre and good (Hutcheson & Sofroniou, 1999). Two factors had eigenvalues over
Kaiser’s criterion of 1, and together explained 73% of the variance.
The items that cluster on the same factors suggest that our hypotheses regarding
evaluations of legally reasonable police use-of-force were accurate, with POLHITOK,
POLESCAP, and POLATTAK cleanly loading on one factor (a legally reasonable factor), and
POLABUSE and POLMURDR cleanly loading on another factor (a legally unreasonable factor).
These findings line up well with the factor loadings Barkan and Cohn (1998) and Johnson and
LEGAL AND COMMUNITY STANDARDS
22
Kuhns (2009) reported in their respective studies, although Johnson and Kuhns labeled their
factors “reasonable” and “excessive.”
Table 1. Factor Analysis of Five Use-of-Force Questions
Legally Reasonable
Legally Unreasonable
POLHITOK
.68
.03
POLESCAP
.87
.18
POLATTAK
.92
-.21
POLABUSE
.34
.68
POLMURDR
-.12
.98
Bearing these findings in mind, and in line with our theoretical structure, we
conceptualize an unreasonable expectation, in the legally reasonable/unreasonable sense, as those
respondents who answer “no” to POLHITOK, POLESCAP, and POLATTAK. Again, while
POLESCAP and POLATTAK are more context-specific than POLHITOK, POLHITOK still
provides valuable insight into the public’s evaluation of police use-of-force within our
conceptualization. As such, and in light of the factor analysis results, POLHITOK is retained in
the legally reasonable factor.
As we are only interested in the factors underlying unreasonable expectations in this
study (again, in the legally reasonable/unreasonable sense), only the items contained within the
legally reasonable factor were used in subsequent analyses. That is, we seek to document the
factors associated with respondents answering “no” to situations in which force would be
deemed legally and professionally reasonable.
Dependent variable operationalization. Given the above conceptualization of the
dependent variable, the dependent variable is operationalized as a summed index of the three
items that load on the legally reasonable factor, which results in a count variable ranging from
zero to three. Descriptive statistics for the resulting variable, titled “Use of Force Perceptions,”
are displayed in Table 2, while descriptive statistics for the underlying items are reported in
LEGAL AND COMMUNITY STANDARDS
23
Table A1 in Appendix A. Further, a figure depicting the conditional ‘and’ responses is included
in Figure A1 of Appendix A.
Just over 58% of respondents have a legally reasonable view of all three use-of-force
questions across twenty-six years of the GSS, while respondents with the most negative
evaluations of legally reasonable force (scoring a three on the count variable) are low enough in
number to be considered a rare event for modeling purposes. The proportion of respondents with
“no” answers to at least one of the questions is 41.71%, while only 6.64% of respondents have a
legally unreasonable view of all three questions regarding use-of-force. The distinction between
the summed scale as we operationalize it, and the raw count per item, both of which are reported
in Table 2, support our operationalization. It is not only the disapproval of any single legally
reasonable use-of-force question which is interesting, but also how respondents conceive of
police use-of-force across the three items.
Of concern to the dependent count variable is that respondents are not conditionally
consistent in how they answer the questions. For instance, we might expect that someone who
would never approve of an officer to strike an adult male (POLHITOK=1) would also never
approve of an officer striking an escaping suspect (POLESCAP=1) or to defend himself from
attack (POLATTAK=1). While most respondents are consistent (all “no” or all “yes”; n=13,492),
not all are. Some of the inconsistency can be attributed to respondents gathering more context in
the specific use-of-force questions, but not all of the conditional inconsistencies can be explained
so optimistically. Some respondents, for example, say they can imagine when a strike would be
reasonable (POLHITOK=0), yet 3.44% of that group says they would not approve of a strike in
the self-defense scenario (POLATTAK=1).
LEGAL AND COMMUNITY STANDARDS
24
There is no way to establish whether this type of discrepancy is simply an error of
question interpretation or some other reason, and attempting to explain why respondents hold that
specific combination of beliefs is outside the scope of the study in any case. However, the
inconsistencies are important because they underline the study’s concern with the difficulty in
aligning the legally reasonable standards with public opinion on police use-of-force. A figure
depicting the conditional responses of respondents to the three retained use-of-force questions is
not depicted here but is available in the study’s appendix. We return to this discussion in more
detail in the findings section of this manuscript.
Independent and control variables. Operationalization details of the independent
variables race, political views, and education, and the controls for sex, income, age, population
of respondent’s place of residence, and year of survey are displayed in Appendix A. Descriptive
statistics for all measures are presented in Table 2 below.
Table 2. Descriptive Statistics
Variable
n
Frequency %
Mean
SD
Range
Individual Dependent Variable
POLHITOK
20,781
.301
.459
0-1
POLESCAP
20,781
271
.444
0-1
POLATTAK
20,781
.010
.230
0-1
Summed Dependent Variable
Use of Force Perceptions
20,781
.671
.928
0-3
Count is 0
12,113
58.29
Count is 1
4773
22.97
Count is 2
2516
12.11
Count is 3
1379
6.64
Independent Variables
Real Income ($1986)
20,781
29,891.
28
31,860.04
$0 -
$155,140
Age
20,735
46.52
17.28
18 - 89
Location Population (in 1000s)
20,781
335.01
1161.48
0 - 8175
LEGAL AND COMMUNITY STANDARDS
25
Race
20,781
1.28
.592
0-2
Black
2843
13.68
Other Race
1533
7.38
White*
16,405
78.94
Sex
20,781
1.55
.496
1-2
Male*
9442
45.44
Female
11,339
54.56
Education
20,746
1.55
1.19
0-4
Less than High School*
2979
14.36
High School
10,779
51.96
Jr. College
1509
7.27
Bachelor’s Degree
3647
17.58
Graduate Degree
1832
8.83
Political Views
20,045
2.09
.787
1-3
Liberal
5404
26.96
Moderate*
7493
37.38
Conservative
7148
35.66
Fully 30% of General Social Survey respondents (POLHITOK) disapprove of a police
officer ever striking a male adult citizen, while 27% endorse a similar view for an officer ever
striking a citizen trying to escape custody. Also, while significantly fewer respondents have a
negative evaluation regarding police use-of-force against a citizen striking a police officer with
their fists, it is notable that nearly one-in-ten respondents offer a negative evaluation of this
scenario. The response pattern is striking given the specific context of the latter situation: police
officers defending themselves against a physical assault.
Negative Binomial Model Specification
The dependent variable, Use of Force Perceptions, is a summed scale of the three
dichotomous items of interest (POLHITOK, POLESCAP, and POLATTAK), and the resulting
outcome variable is an event count that attains only non-negative, whole integer values between
0 and 3. While a Poisson distribution is more commonly used to model such event counts, that
statistical analysis requires equidispersion of the mean and variance. In the case of the data being
LEGAL AND COMMUNITY STANDARDS
26
analyzed here, the Pearson goodness-of-fit criterion indicates the distribution significantly differs
from the Poisson distribution, as the conditional variance is greater than the conditional mean. In
such cases, there is agreement among researchers that a negative binomial regression is an
appropriate alternative (see Hible, 2011; Dowler & Zawilski, 2007).
The dependent variable displays a high proportion of zeroes due to the decision to code
respondent answers to the underlying items as zeroes for “yes” answers, and ones for “no”
answers. Raw GSS data for these questions uses (no=2, yes =1), and our transformation
introduces zeroes and ones, but this is an operational decision that does not affect the underlying
process, and model results are robust to either coding scheme. In all three underlying questions,
the majority of answers are the legally reasonable “yes,” and so in the summed count dependent
variable there is a substantial proportion of zeroes. While this may indicate that a zero-inflated
negative binomial (ZINB) model would be the ideal choice, in this case it is not appropriate. The
ZINB model requires a theoretical interpretation in which there are two or more ‘kinds’ of zeroes
in the data (Greene, 1994). Such a theoretical interpretation does not fit the underlying data
studied here.
The heavy proportion of zeroes in a dependent variable is not enough on its own to
require a ZINB model. There must also be multiple processes leading to the zeroes, such as
individuals who have zero probability of a count higher than zero. A ZINB model specification is
appropriate, for example, when the outcome of interest does not occur (zero count), can occur
but not be recorded (zero inflation), or cannot occur in some cases (zero inflation). In outcomes
like this, there is process heterogeneity that artificially inflates the zero count, which if modeled
with Poisson or NB distributions leads to poor estimation.
LEGAL AND COMMUNITY STANDARDS
27
In this study’s dependent variable, there is only one process for a respondent’s answer to
be “no” (and thus coded as a one), or “yes” (and thus coded as a zero). No respondent is
structurally incapable of answering “yes” or “no.” There are not multiple processes leading to an
inflation of zeroes, as “zero” is a coding choice for cases, rather than a physical count where zero
is the absence of an outcome of interest. In other words, there are a lot of zeros, but they are not
inflated zeros and could just as easily be “deflated” (yes=1, no = 0) with a different coding
decision.
Negative binomial regression is used here to investigate our fundamental research
question – how often does the ‘event’ of a respondent answering the use-of-force questions
contained within the GSS occur in what has been conceptualized here as a legally unreasonable
way? Negative binomial regression allows us to assess how a one-unit change in each of the
independent variables affects the likelihood of a negative evaluation of legally reasonable police
use-of-force as conceptualized here. The results of the analysis appear in Table 3.
Table 3. Negative Binomial Model: Number of Legally Unreasonable Answers to GSS Use-
of-Force Questions
Use of Force Perceptions (Count)
IRR (Std. Err.)
[95% Conf. Interval]
Female
1.434(.028)**
1.380
1.491
Race
Black
1.908(.046)**
1.189
2.000
Other Race
1.887(.059)**
1.773
2.007
Political Ideology
Liberal
1.070(.025)*
1.023
1.121
Conservative
.862(.020)**
.823
.901
Formal Education
High School
.707(.018)**
.672
.744
Jr. College
.662(.028)**
.609
.720
Bachelor’s Degree
.539(.019)**
.503
.578
Graduate Degree
.598(.027)**
.548
.653
Age
1.000(.0005)
.999
1.001
LEGAL AND COMMUNITY STANDARDS
28
Population
1.000(.000)**
1.00003
1.00006
Real Income ($1986)
.9999(.000)**
.999996
.999998
Year
1.010(.001)**
1.008
1.013
Constant
6.07e-10
(1.32e-09)**
8.53e-12
4.32e-08
Pseudo R2
0.0542
LR chi2
2399.93
alpha
0.237 (0.021)
0.199
0.281
Log-likelihood
-20938.889
Observations
19,972
Note: reference categories not displayed for education (less than high school), sex (male), race (white),
and political views (moderate); Significance levels at * p £ .01 and ** p £ .001
Findings
The general picture emerging from the analysis is in line with previous studies suggesting
that demographic factors condition the likelihood of respondents’ views regarding police use-of-
force. The chi-squared statistic reported in Table 3 indicates that all iterations of the complete
regression model are significant at the .01 level. Statistical significance was accepted at the .01
level for independent variables. All independent variables have a significant effect on the count
outcome in the direction predicted by our research hypotheses.
We found strong support for hypothesis 1, which pertains to race and minority ethnicity.
While other variables in the model are held constant, black respondents are expected to have a
rate 1.908 times greater than Non-Hispanic white respondents for disapproving of legally
reasonable use-of-force, while non-white, non-black minority group respondents are predicted to
have a rate 1.887 times greater.
As in previous studies, the results of this analysis confirm significant support for
ideological effects on citizen views of police use-of-force (hypothesis 2). Respondents who
identify as having liberal political views are expected to disapprove of legally reasonable use-of-
LEGAL AND COMMUNITY STANDARDS
29
force at a rate 1.070 times greater than those who identify as moderate, while conservative
respondents would be expected to have a decreased rate of disapproval by a factor of 0.862.
The data demonstrate the effect of formal educational attainment on respondents’ views
of police use-of-force (hypothesis 3). The results reveal that there are significant effects across
education levels, though not in a perfectly linear fashion. The model makes use of respondents
with a less than high school education as the reference category. The results indicate that as
respondents increase in their education level through a bachelor degree, there is a steady
decrease in disapproval of justified uses of force by police. All else equal, when compared to
those without a high school diploma, an increase in educational level reduces disapproval by a
factor of 0.707 (high school), 0.662 (junior college), and 0.539 (bachelor degree). While
graduate-level education still produces a significant decrease in the expected rate of legally
unreasonable answers by a factor of 0.598, when compared to the least educated respondent
class, it slightly reverses the trend observed through the bachelor degree level. Given the
difficulty in directly interpreting exponentiated coefficients along with the small absolute
differences observed suggest that the most defensible interpretation of the effect of formal
education is that it generally decreases legally unreasonable opinion on police use-of-force.
Except for respondents’ age, all control variables (sex, income, size of city wherein one
resides, and year of the survey) were found to be statistically significant. The model indicates
that when compared to men, women are more likely to disapprove of legally reasonable use-of-
force scenarios (IRR=1.434). The effects of the control variables for income, city population, and
time (year) are all significant though minimal. As real income increases ($1986), the model
predicts a decrease in use-of-force disapproval (IRR=.999). As population size increases, the
model expects a respondent to increase their disapproval of legally reasonable use-of-force
LEGAL AND COMMUNITY STANDARDS
30
scenarios (IRR=1.00). Finally, the model shows that with all else held constant, over time
disapproval of legally reasonable use-of-force increases (IRR=1.01).
Discussion
Our results indicate that a sizeable portion of the American public tends to express
disapproval of legally reasonable police use-of-force. It should be kept in mind that these results
may not generalize to attitudes for all forms of police use-of-force. Striking a citizen is but one
form of force, and police striking citizens may not align with attitudes toward police use of
physical force that might include the use of a baton, use of less-than-lethal weapons such as a
taser, or use of lethal force.
With the above noteworthy reservations in mind, the results of our research correspond
with past scholarship in that race and minority ethnicity appear to be the most substantial
influences on citizen assessments of police use-of-force (Arthur & Case, 1994; Brown &
Benedict, 2002; Halim & Stiles, 2001; Hurst et al., 2000; Roche & Roux, 2017; Thompson &
Lee, 2004; Tuch & Weitzer, 1997). Black and other minority respondents were much more likely
in comparison to non-Hispanic white respondents to have what has been conceptualized as
legally unreasonable expectations of police use-of-force. This finding held even when controlling
for sex, age, political views, and socioeconomic status indicators. Scholars have proposed that,
specifically for African-Americans, this effect is likely due to historical influences (Johnson &
Kuhns, 2009). Hurwitz and Peffley (2005) found support for the hypothesis that blacks’ global
attitudes for the evaluation of police interactions are more negative than for whites. They state in
this regard, “Given the history of racial bias in the system, African Americans should be more
vigilant to signs of discrimination in encounters between police and…citizens” (p. 767).
LEGAL AND COMMUNITY STANDARDS
31
Small (1983) reported the same phenomenon with black subjects having a negative global
attitude toward the police. He found this negative perception of the police could not be explained
based solely on personal experiences, but rather by stories from family, friends, and the media
which reinforced the larger negative narrative. Small’s findings may apply to the results of this
study, as black and other minority respondents expressed higher levels of disapproval of what
prima facie are broadly considered legally reasonable circumstances for the use of force by
police officers.
Although the literature on the expected effect of political identification and normative
evaluations of police use-of-force is not extensive, it is consistent; accordingly, we hypothesized
that respondents with liberal political views would be more likely to have more negative, and
conservatives less negative, evaluations of legally reasonable police uses of force when
compared to self-identified moderates. The results here provide preliminary evidence that even
when other demographic characteristics are held constant, a respondent’s self-reported ideology
has a significant effect on their views on police use-of-force issues. As with race and ethnicity,
the ideological divides between conservative values and liberal values (Braithwaite, 1998; Davis
& Silver, 2004; Graham, Haidt, & Nosek, 2009) also influence evaluations of police use-of-
force.
For policymakers, one critical takeaway of these results is that negative evaluations of
legally reasonable police use-of-force are heavily influenced by demographic factors outside the
control of police officers and police administrators. Exploring similar ground, Braga et al. (2014)
likewise found that social contextual factors, such as citizen political ideology, had a significant
influence on appraising the fairness of officers’ actions. If true, and perceptions of police use-of-
force are driven by unalterable factors beyond the control of the police, the police and those
LEGAL AND COMMUNITY STANDARDS
32
citizens with arguably unreasonable expectations for police conduct may be caught in a situation
perpetuating mistrust.
One may argue that the actions of officers in instances of use-of-force can influence
citizens’ perceptions, and that the preceding analysis leads to an overly cynical conclusion.
Perceptions of procedural justice (i.e., the evenhandedness and fairness of police procedures) are
argued to be the primary antecedent to police legitimacy (Wolfe et al., 2016). Perhaps some
cynicism is warranted, but it remains an open question as to the full degree to which police
behaviors affect the public’s subjective perceptions of procedural justice (Nagin & Telep, 2017).
This lack of clarity is because procedural justice research has almost exclusively relied on citizen
survey data in the documentation of the phenomena. In a recent report, the National Academies
of Sciences, Engineering, and Medicine (2018) observed that only one study (Worden &
McLean, 2014) had assessed the association between subjective perceptions of procedurally just
treatment and actual police treatment. Worden and McLean found that citizens’ subjective
perceptions of procedural justice and officers’ actual procedural justice-related behavior are not
mere reflections of one another. The authors concluded from their observations and data
collected that it is “imperative to draw a sharp distinction between procedural justice as citizens’
subjective experience and procedural justice as officers’ overt behavior. They are different
phenomena…” (p. 18).
Worden and McLean argue that whether officers used physical force was more important
than how they used force in affecting citizen assessment of procedural justice. They reported that
the how largely did not matter mainly because citizen assessments were shaped primarily by
influences beyond officers’ control, such as citizens’ background, situational context, citizens’
race, age, and gender. Further, findings drawn from panel data and other types of research into
LEGAL AND COMMUNITY STANDARDS
33
citizen attitude formation regarding their perception of the police reinforce the overall idea that
citizens’ subjective evaluations of a police interaction are shaped more by their prior attitudes
than contemporaneous police interactions (Sargeant et al., 2018; Worden & McLean, 2014).
Several explanations for both the results of this study and the Worden and McLean
(2014) study are available. First, some members of the public may make judgments about police
use-of-force based mostly on emotions (Shane & Swenson, 2019). Indeed, when controversies
over police use-of-force arise, they are often highly emotionally charged, and it is quite well
understood that evaluations and actions based on emotion (especially negative emotion) can
impair the kind of rational decision-making (Kahneman, 2003) upon which legal analysis is
founded. When emotionality is high, reasoned legal analysis is unlikely to predominate. The
inconsistent inter-item patterns we highlight elsewhere are empirical evidence that full legal
rationality is not a reality. An assumption of legal rationality also assumes that members of the
public have sufficient, capable knowledge of constitutional law and the legal system. Research
on this matter tends to undermine this assumption (Jamieson & Hennessy, 2006), and may, in
part, help explain the observed reduction in disapproval of legally reasonable use-of-force as
formal educational attainment level increased. That is, as education increases, exposure to
constitutional law and the legal system, as well as exposure to critical thinking, increases, thus
likely providing a more legally-informed view of police use-of-force situations.
Second, there may be an evaluative mismatch with use-of-force incidents due to a
knowledge and experience gap when it comes to the human factors typically involved in police
use-of-force events (Klinger & Brunson, 2009). While some argue that complaints against police
officers for excessive use-of-force do not receive the scrutiny and legal consequences if
sustained upon independent investigation that they deserve (Lacks, 2008), others argue with
LEGAL AND COMMUNITY STANDARDS
34
equal force that at times citizens hold officers to an impossible, superhuman standard of conduct
(Blair et al., 2011). To better understand what are ‘human’ and ‘superhuman’ expectations of a
reasonable officer, researchers have begun to study human factors (Hope, 2016; Kleider-Offutt,
Clevinger, & Bond, 2016; Nieuwenhuys, Savelsbergh, & Oudejans, 2015) that place limitations
on what one can reasonably expect from police officers involved in stressful situations.
A host of psychophysiological reactions (i.e., perceptual distortions, memory distortions,
and reaction time) have an impact upon any human being’s ability to gather information on, and
react to, critical situations. Such research is yet at a nascent stage, but several studies do shed
some light on what can reasonably be expected of officers. For example, Klinger and Brunson
(2009) showed a decade ago that the majority of officers in their sample who had been involved
in a shooting experienced at least two noteworthy perceptual distortions during the incidents in
which they were involved. Most often these perceptual distortions were those of auditory
blunting and tunnel vision. These natural distortions are subjective and cannot be ‘seen’ by
anyone other than the officer experiencing them. This divergence of ‘felt subjective experience’
and ‘seen objective display’ may result in perceptions of unreasonable force, when in fact it is a
reasonable human reaction experienced by people acting under stress. A long, but highly relevant
passage from Klinger and Brunson describes precisely how such pertinent distortions can arise in
police use-of-force situations (p. 134-135):
“The reality of a given situation might be, for example, that an officer is shouting
for her partner not to shoot because she has properly identified the pistol held by a
robbery suspect standing in a dark alley to be a toy. Because her partner is
experiencing auditory occlusion, however, he does not hear her shouts and takes
the suspect under fire because he believes the toy to be a real gun. Similarly, the
shooting officer might continue to fire after the toy gun has fallen from the suspect’s
hand because tunnel vision has restricted his visual field to the suspect’s torso,
where he is aiming his shots. Continuing with our hypothetical, a civilian witness
with a video-equipped cell phone observes and records the event; then the witness
goes on to present both testimonial and electronic evidence demonstrating that (1)
LEGAL AND COMMUNITY STANDARDS
35
the suspect did not have a real firearm, (2) the partner officer loudly told the shooter
not to fire, and (3) the officer who took the suspect under fire…continued to fire
after the suspect dropped the object from his hand…Based on the actual threat…it
would seem clear that he should not have fired…but things are not so clear if we
judge the shooter’s actions from his perspective.”
Officers frequently experience sensory distortions during use-of-force incidents (Klinger
& Brunson, 2009), but these normal limits of human performance are not well known and can
result in the perception of unreasonable use of physical force. As such, police officers and some
members of the public are likely to evaluate use-of-force incidents very differently (Alpert &
Smith, 1994; Herbert, 2006; Mourtgos et al., 2019). A deeper understanding of what officers
experience during a use-of-force incident may lead to better training of the police and the legal
profession (prosecutors and attorneys for the defense alike) and help reduce tension between the
police and the public (Klinger & Brunson, 2009).
Addressing Inconsistent Public Opinion
Our findings indicate that many individuals hold unreasonably disapproving attitudes
concerning police use-of-force issues. Further, the robustness of legally unreasonable public
opinion on police use-of-force is alarming, with 41.7% (n=8,668) of respondents having at least
one legally unreasonable response to the three GSS items analyzed here. Looking past the inter-
item inconsistencies discussed below, if respondent answers are accurate reflections of their
genuinely held belief, the fact that 30.06% of respondents would never approve of a police
officer striking a suspect casts significant doubt on the ability for many citizens to provide an
informed view of what is reasonable in police use-of-force situations. Just over 27% also did not
approve of striking a citizen who was attempting to escape custody, and perhaps most
concerning, 9.98% of respondents did not approve of officers striking a citizen even when that
citizen is physically attacking the officer.
LEGAL AND COMMUNITY STANDARDS
36
However, the issue of inter-item consistency is not ignorable and raises concern beyond
the percentages of legally unreasonable views at the level of the survey items. In some cases, this
inconsistency can be construed as a reason for optimism. For example, of those who report they
would never approve of a police officer striking an adult male, and maintain that disapproval for
striking a suspect escaping custody (n=3,328), 58.56% (n=1,949) say they would approve of an
officer striking an adult male who was attacking the officer. The change from never approve to
approve conditionally might be explained by assuming that with more pointed context, some
subset of initially legally unreasonable respondents is willing to change their mind. Such an
explanation gives hope that our later recommendation for increased public education and active
engagement might reap at least some benefit.
That hopeful conclusion is tempered by both the consistency of legally unreasonable
opinion by a minority of respondents, as well as the inconsistencies that do not break so clearly
for the optimistic side of interpretation. First, there are subsets of respondents that are
consistently unreasonable. As indicated by the highest count in our dependent variable, some
6.64% of respondents answer ‘no’ to all three underlying questions. Taken on its face, and
putting aside the possibility of measurement or respondent error, this is a subset of the population
that appears dedicated to a legally unreasonable interpretation of police use-of-force.
A second type of inconsistency resists attempts to parse logically. What to make of the
3.44% of respondents (n=500) who can generally think of times where it would be okay for an
officer to strike an adult male (POLHITOK=0), but who would not approve of an officer striking
in self-defense (POLATTAK=1)? Response or measurement error seems a likely candidate for at
least some of that type of inconsistency but is unlikely to explain all of it, and the practical public
LEGAL AND COMMUNITY STANDARDS
37
policy response is unclear. A figure displaying the inter-item consistency for the individual
survey items underlying the summed dependent variable is available in the study’s appendix.
Generally, the mix of mostly consistently legally reasonable (n=12,113), consistently
legally unreasonable (n=1,379), and inconsistent (n=7,289) public opinion is evidence in favor of
both the study’s importance and its dependent variable operationalization. First, there is tension
at every level of public opinion on police use-of-force measured here, which matches the
national climate of tension generally found around police use-of-force. The response variability
shows that public opinion provides irregular ground for structuring police use-of-force, and
should provoke more nuanced discussion and investigation into the underlying formation of
those opinions.
Second, the variability and inconsistency suggest that a count variable, despite its
limitations, is an appropriate operationalization for the dependent variable used in our analysis.
Analyzing a single item general response (POLHITOK) gives little understanding of the nuance,
while analysis of the most facially defensible type of police force (POLATTAK) risks
overlooking the relatively broad subset of American’s who are discomfited by the very idea of
governmental violence no matter the context. Neither is satisfactory, and our count variable
solution is far from perfect. However, to the degree that all research designs are choices among
imperfect alternatives, our choice falls somewhere short of perfection while remaining defensible
and tractable. Whether the results are useful must be left to the passing of time.
Addressing the Findings
Our findings raise the question of what can be done to increase police legitimacy when it
comes to the public’s evaluation of police use-of-force if the public’s evaluations are far from
LEGAL AND COMMUNITY STANDARDS
38
aligning with the legal and professional training standards police officers are expected to uphold.
We make two suggestions centered on public education and active engagement with citizens.
Some agencies have utilized programs such as citizen police academies to help members
of the public learn about police practices and use-of-force – both the legalities at play and the
human limitations involved. However, the people that attend these academies are most typically
already supportive of the police (Engel & Smith, 2009). More attention should be paid to
reaching those who are inclined to have unfavorable views of the police such as low
socioeconomic status individuals, women, and racial and ethnic minority community members.
These groups were shown to be more likely to disapprove of legally reasonable use-of-force.
Indeed, some educational opportunities have influenced ardent critics of police use-of-force to
moderate their views substantially (Chiaramonte, 2015). However, these programs must be able
to scale up effectively. Teaching a few citizen police academies a year, or providing isolated use-
of-force education opportunities, are actions unlikely to have a demonstrable effect in medium
and large cities. Instead, educational programs on the scale of public ad campaigns should be
piloted alongside the more direct and substantive public education that takes place in citizen
academies.
Engel and Smith (2009) also point out that police agencies generally have not addressed
the need to educate their citizens concerning the complexities of police use-of-force. In the
absence of such information, much of the public is left to form views of police use-of-force
through a frame mediated by popular culture depictions. With most citizens never having dealt
with subduing a resisting individual, or having to control a chaotic and dangerous situation
“under conditions of fear, panic, and chaos” (Engel & Smith, 2009, p. 141), they are often left to
evaluate police actions based on television portrayals, movies, and the vicarious experiences of
LEGAL AND COMMUNITY STANDARDS
39
others with whom they come in frequent contact such as family, friends and workplace
acquaintances (Dowler & Zawilski, 2007; Eschholz, Blackwell, Gertz, & Chiricos; Small, 1983).
With that background in mind, another educational possibility has arisen with the rapid
and widespread adoption of body-worn cameras (BWCs) in police departments. While recent
work has highlighted the lack of consistent research findings on the impact of BWCs on use-of-
force (Lum, Stoltz, Koper, & Scherer, 2019), as well as the impact of BWCs on vulnerable
populations (Adams & Mastracci, 2017) and burnout in officers themselves (Adams &
Mastracci, 2018), one positive outcome may be the ability for police departments to produce and
distribute realistic videos of police use-of-force culled from BWC footage. Such footage may
help combat the often unrealistic views portrayed in popular entertainment.
The recently developed television show “Body Cam” on the Discovery Channel (Body
Cam, 2018) is attempting to do precisely that, but some caution is warranted. The ‘reality police’
television show concept is not a novel invention, and some research indicates that such television
programs may inadvertently increase the “racial divide” in citizen attitudes toward the police
(Eschholz et al., 2002). To forestall this outcome, and to utilize such shows and BWC footage as
educational tools, it is imperative that legal process-based explanations and current information
regarding human limitations are included in the interpretation of the incidents taking place. Much
of the CIT training noted earlier in modern police training and academy settings entail just such
use of BWC video archives.
Second, if progress is to be made in police-community relations, the evidence and
analysis presented here suggest a two-pronged approach to relationship building between the
police and the public is required. While continued progress and attention to guardianship-
oriented policing as advocated in the President's Task Force on 21st Century Policing are
LEGAL AND COMMUNITY STANDARDS
40
necessary, so too is an improvement in public understanding and knowledge about why police
take use-of-force actions, and what situational factors typically make it necessary and reasonable
to do so. That is, these issues of police legitimacy cannot fall solely on the shoulders of the
police (Mourtgos et al., 2019). It takes an engaged citizenry to ensure a robust democratic
society, and mutual trust is necessary for sustaining democratic governance (Yang, 2005). The
public at-large should engage constructively with their local police departments in their
respective communities to understand more fully the police culture and the challenges facing
police officers. Unfortunately, citizen interest in participation in police engagement programs is
too often underwhelming (Herbert, 2006), particularly in communities where tensions and
historical events have led to animosity. A genuine commitment to improving police-community
relations requires equal involvement on the part of the police and the citizens they serve.
A third possible policy approach is the alteration of the legal standards for police use-of-
force. Recently, the State of California has considered legislation that would change the use-of-
force standard from ‘reasonable’ to ‘necessary’ (Bollag & Thompson, 2018). Under this
standard, officers would only be authorized to use deadly force when it is ‘necessary,’ rather than
when it is deemed ‘reasonable’ under the circumstances presenting themselves to an officer. It is
unclear how this standard would work based on the discussion above regarding evolving
evidence on human performance limitations. If the ‘necessary’ standard were enacted, an officer
faced with a subject pointing a realistic looking toy gun at them (as in the hypothetical situation
presented earlier) could be criminally charged for using unlawful force by shooting that person.
This solution may lead to an unworkable standard for officers as “the decisions that officers
make about firing their weapons will frequently be based on perceptions of the situation that do
not enjoy a one-to-one correspondence with objective reality” (Klinger & Brunson, 2009, p.
LEGAL AND COMMUNITY STANDARDS
41
134). Nonetheless, it is an option that forms part of the “national conversation on reforming
police practices by evaluating the current law on police use of deadly force” (Lee, 2018, p. 629).
Limitations
The findings presented here contribute importantly to the ongoing national debate
regarding police use-of-force, but the study is not without limitations. First, as in previous
studies using the General Social Survey data, the racial characteristics of respondents are
challenging to operationalize. While the survey has made strides in recent years in going beyond
black/white/other classifications, this study is forced to use the older survey race scheme to
maintain consistency over the 1990 to 2018 period. In light of this limitation, few conclusions
can be drawn for the impact of those characterized as having an “other race” categorization
despite the relatively large effect documented in the regression model summary statistics.
Second, this study suffers from the usual validity threats found in survey-based research,
and despite including controls for the generally increasing cynical national mood developing
throughout the years surveyed, there are likely unexamined confounding variables which could
be included in such a longitudinal study. Confounding variables may include the influence of
media stories about past or recent police-community conflicts, jurisdictional idiosyncrasies
within and across locations, and environmental factors (i.e., jurisdictional policing styles and
crime rates). Recognizing the limitations of using survey data to examine this topic, experimental
methods have been used effectively to investigate the dynamics associated with public
perception of police use-of-force (Braga et al., 2014; Patil, 2018), and future research should
continue to leverage the power of multiple methods to investigate the justice perceptions of the
general public.
LEGAL AND COMMUNITY STANDARDS
42
Third, one limitation of our operationalization of the dependent variable as a count
variable is that to some degree it assumes a stability, or logic, within respondent answers to the
three questions which we conceptualize as having a ‘legally reasonable’ answer (“yes”). That
assumption of validity can be undercut in two ways. First, there is limited enough covariance
between the three to suggest a lack of opinion stability between them. Second, and more difficult
to intuit, are the portions of the public with directly conflicting inter-item scores. The presence of
conflicting inter-item scores is discussed in much more detail above in the discussion section of
this manuscript, and this limitation is mitigated by the relatively small proportion of observations
which display this inconsistency and is partially addressed statistically in our factoring strategy.
The final limitation of this study is an artifact of large national-based surveys, in that how
the public perceives police use-of-force is contextually dependent. We posit there is a legal and
normatively reasonable answer to the three questions investigated here. However, researchers
cannot realistically expect respondents to respond tabula rasa, without experience or the
personal context which has formed the values and beliefs underlying their views of police,
violence, or police use-of-force. Several experimental researchers (Patil, 2018; Testa & Dietrich,
2017; Culhane Boman, & Schweitzer, 2016) have demonstrated the substantial influence of
recency on how the public perceives the justifiability of police use-of-force. Given these effects,
some respondents who are characterized within our study as disapproving of reasonable police
use-of-force are likely to be responding to a specific incident within their own experience or
perception, or vicarious account of a family member or acquaintance, rather than the
universalized phrasing of GSS questions. This limitation is outside the scope of this study, but is
a consideration for future research aiming to understand public perceptions of police behavior.
Conclusion
LEGAL AND COMMUNITY STANDARDS
43
Policing American communities is a demanding task. This study shows that a sizeable
portion of the American public disapproves of police use-of-force in situations in which such
action would be legally and professionally reasonable. The tension between the legal standards
on one side, and the evolving and sometimes legally unreasonable expectations of the public on
the other, makes the job of policing extremely difficult for front-line police officers and their
administrators. Continuing to professionalize law enforcement in the United States is vital,
including raising performance standards, decreasing incidents of use-of-force where possible,
and holding officers accountable when appropriate. However, that list of goals reflects normative
goals already in place across much of the country, and the findings on citizen attitude formation
reported here indicate a growing gap between legal structures, professional codes of conduct, and
the public regarding police use-of-force. Reducing the space between legal expectations on the
one hand, and community expectations on the other, is an increasingly salient public policy
priority.
LEGAL AND COMMUNITY STANDARDS
44
References
Adams, I., & Mastracci, S. (2018). Police body-worn cameras: Effects on officers’ burnout and
perceived organizational support. Police Quarterly, online edition:
https://doi.org/10.1177/1098611118783987
Adams, I., & Mastracci, S. (2017). Visibility is a trap: The ethics of police body-worn cameras
and control. Administrative Theory & Praxis, 39, 313-328.
Adams, K. (1999). What we know about police use of force - Use of force by police: Overview of
national and local data. US Department of Justice, Bureau of Justice Statistics,
Washington, DC.
Adler, B. (2019, May 24). In California, Agreement On New Rules For When Police Can Use
Deadly Force. NPR.Org. Retrieved from
https://www.npr.org/2019/05/24/726500537/in-california-agreement-on-new-rules-
for-when-police-can-use-deadly-force
Alpert, G. P., & Dunham, R. G. (2004). Understanding police use of force: Officers, suspects,
and reciprocity. New York: Cambridge University Press.
Alpert, G. P., & Smith, W.C. (1994). How reasonable is the reasonable man: Police and
excessive force. Journal of Criminal Law and Criminology, 85, 481-501.
Arthur, J. A., & Case, C. E. (1994). Race, class and support for police use of force. Crime, Law,
and Social Change, 21, 167-182.
Barkan, S. E., & Cohn, S. F. (1998). Racial prejudice and support by whites for police use of
force: A research note. Justice Quarterly, 15, 743-753.
LEGAL AND COMMUNITY STANDARDS
45
Bartels, E. C., & Silverman, E. B. (2005). An exploratory study of the New York City civilian
complaint review board mediation program. Policing - An International Journal of Police
Strategies & Management, 28, 619-630.
Beetham, D. (1991). The legitimation of power. New Jersey: Humanities Press International, Inc.
Blair, P. J., Pollock, J., Montague, D., Nichols, T., Curnutt, J., & Burns, D. (2011).
Reasonableness and reaction time. Police Quarterly, 14, 323-343.
Body Cam. (2018). Discovery, Inc. Retrieved from http://www.imdb.com/title/tt9425082/
Bottoms, A., & Tankebe, J. (2012). Beyond procedural justice: A dialogic approach to
legitimacy in criminal justice. The Journal of Criminal Law & Criminology, 102, 119-
170.
Bollag, S., & Thompson, D. (2018, June 19). California advances biggest US change to police
use of force. Retrieved from http://www.startribune.com/california-advances-biggest-us-
change-to-police-use-of-force/485979461/
Braithewait, V. (1998). The value orientations underlying liberalism-conservatism. Personality
and Individual Differences, 25(3), 575-589.
Braga, A. A., Winship, C., Tyler, T. R., Fagan, J., & Meares, T. L. (2014). The salience of social
contextual factors in appraisals of police interactions with citizens: A randominzed
factorial experiment. Journal of Quantitative Criminology, 30, 599-627.
Brown, B., & Benedict, Wm R. (2002). Perceptions of the police: Past findings, methodological
issues, conceptual issues and policy implications. Policing: An International Journal of
Police Strategies & Management, 25, 543-580.
Brown, K., & Coulter, P. B. (1983). Subjective and objective measures of police service
delivery. Public Administration Review, 43, 50-58.
LEGAL AND COMMUNITY STANDARDS
46
Brunson, R. K., & Weitzer, R. (2011). Negotiating unwelcome police encounters: The
intergenerational transmission of conduct norms. Journal of Contemporary Ethnography,
40(4), 425-456.
Cao, L., Frank, J., & Cullen, F. T. (1996). Race, community context, and confidence in the
police. American Journal of Police, 15, 3-22.
Carter, J. S., & Corra, M. (2016). Racial resentment and attitudes toward the use of force by
police: An over-time trend analysis. Sociological Inquiry, 86, 492-511.
Chamlin, M. B., & Cochran, J. K. (2006). Economic inequality, legitimacy, and cross-national
homicide rates. Homicide Studies, 10, 231-252.
Chermak, S., McGarrell, E. F., & Weiss, A. (2001). Citizens' perceptions of aggressive traffic
enforcement stratgies. Justice Quarterly, 18, 365-391.
Chiaramonte, P. (2015, February 15). Shoot or don’t shoot: Police scenarios prove eye-opening
for civil rights leaders. Retrieved from https://www.foxnews.com/us/shoot-or-dont-shoot-
police-scenarios-prove-eye-opening-for-civil-rights-leaders
Claud, I. L. (1966). Collective legitimization as a political function of the United Nations.
International Organizations, 20, 367-379.
Coicaud, J. (2007). Legitimacy and politlics: A contribution to the study of political right and
political responsibility. United Kingdom: Cambridge University Press.
Culhane, S. E., Boman IV, J. H., & Schweitzer, K. (2016). Public perceptions of the justifiability
of police shootings: The role of body cameras in a pre- and post-Ferguson experiment.
Police Quarterly, 19, 251-274.
Cullen, F. T., Fisher, B. S., & Applegate, B. K. (2000). Public opinion about punishment and
corrections. Crime and Justice, 27, 1-79.
LEGAL AND COMMUNITY STANDARDS
47
Davis, D. W., & Silver, B. D. (2004). Civil liberties vs. security: Public opinion in the context of
the terrorist attacks on America. American Journal of Political Science, 48(1), 28-46.
Dowler, K., & Zawilski, V. (2007). Public perceptions of police misconduct and discrimination:
Examining the impact of media consumption. Journal of Criminal Justice, 35, 193-203.
Eith, C., & Durose, M. R. (2011). Contacts between police and the public, 2008. U.S.
Department of Justice, Office of Justice Programs. Washington, DC: Bureau of Justice
Statistics.
Engel, R. (2005). Citizens’ perceptions of distributive and procedural justice during traffic stops
with police. Journal of Research in Crime and Delinquency, 42(4), 445-481.
Engel, R. S., & Smith, M.R. (2009). Perceptual distortion and reasonableness during police
shootings: Law, legitimacy, and future research. Criminology & Public Policy, 8, 141-
151.
Eschholz, S., Blackwell, B. S., Gertz, M., & Chiricos, T. (2002). Race and attitudes toward the
police: Assessing the effects of watching “reality” police programs. Journal of criminal
justice, 30(4), 327-341.
FBI. (2017). ucr.fbi.gov. Retrieved from Law enforcement officers killed and assaulted 2016:
https://ucr.fbi.gov/leoka/2016
Field, A. (2013). Discovering statistics using IBM SPSS Statistics: And sex and drugs and rock
'n' roll, (4th ed.). London: Sage.
Fine, A. D., Rowan, Z., & Simmons, C. (2019). Do politics Trump race in determining America's
youths' perceptions of law enforcement? Journal of Criminal Justice.
Fiske, S. T., & Taylor, S. E. (1991). Social Cognition, (2nd ed.). New York: McGraw-Hill.
LEGAL AND COMMUNITY STANDARDS
48
Fogel, D. (1987). The investigatin and disciplining of police misconduct: A compartive view -
London, Paris, Chicago. Police Studies International Review of Police Development, 10,
1-15.
Geller, W., & Scott, M. (1992). Deadly force: What we know. Washington, DC: Police Executive
Research Forum.
Gilley, B. (2006). The meaning and measure of state legitimacy: Results for 72 countries.
European Journal of Political Research, 45, 499-525.
Graham v. Conner, 490 U.S. 386 (1989).
Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets of
moral foundations. Journal of Personality and Social Psychology, (96)5, 1029-1046.
Greene, W. H. (1994). Accounting for Excess Zeros and Sample Selection in Poisson and
Negative Binomial Regression Models (SSRN Scholarly Paper No. ID 1293115).
Retrieved from Social Science Research Network website:
https://papers.ssrn.com/abstract=1293115
Hadden, S. E. (2001). Slave Patrols. Massachusetts: Harvard University Press.
Halim, S., & Stiles, B. L. (2001). Differential support for police use of force, the death penalty,
and perceived harshness of the court. Criminal Justice and Behavior, 28, 3-23.
Herbert, S. (2006). Citizens, cops, and power – Recognizing the limits of community. Illinois:
The University of Chicago Press.
Hilbe, J. M. (2011). Negative binomial regression. New York: Cambridge University Press.
Honig, A., & Lewinski, W. J. (2008). A survey of the research on human factors related to lethal
force encounters: Implications for law enforcement training, tactics, and testimony. Law
Enforcment Executive Forum, 8, 129-152.
LEGAL AND COMMUNITY STANDARDS
49
Hope, L. (2016). Evaluating the effects of stress and fatigue on police officer response and recall:
A challenge for research, training, practice and policy. Journal of Applied Research in
Memory and Cognition, 5, 239-245.
Huang, W., & Vaughn, M. S. (1996). Support and confidence: Public attitudes toward the police.
In Flanagan, T. J., and Longmire, D. R. (Eds.), Americans View Crime and Justice: A
National Public Opinion Survey (pp. 31-45). New York: Sage Publications.
Hurst, Y. G., Frank, J., & Browning, S. L. (2000). The attitudes of juveniles toward the police: A
comparison of white and black youth. Policing – An International Journal of Police
Stragetgies and Management, 23, 37-53.
Hurwitz, J., & Peffley, M. (2005). Explaining the great racial divide: Perceptions of fairness in
the U.S. criminal justice system. Journal of Politics, 67, 762-783.
Hutcheson, G. D., & Sofroniou, N. (1999). The multivariate social scientist: Introductory
statistics using generalized linear models. Sage.
International Association of Chiefs of Police (IACP). (2006). Use of force: Concepts and issues
paper 06. Alexandria, VA: IACP National Law Enforcement Policy Center.
Jamieson, K. H., & Hennessy, M. (2006). Public understanding of and support for the courts:
Survey results. The Georgetown Law Journal, 95(4), 899-902.
Johnson, D., & Kuhns, J. B. (2009). Striking out: Race and support for police use of force.
Justice Quarterly, 26, 592-623.
Kaiser, H. F. (1974). An index of factorial simplicity. Psychometrika, 39(1), 31-36.
Kleider-Offutt, H. M., Clevinger, A. M., & Bond, A. D. (2016). Working memory and cognitive
load in the legal system: Influences on police shooting decisions, interrogation and jury
decisions. Journal of Applied Research in Memory and Cognition, 5, 426-433.
LEGAL AND COMMUNITY STANDARDS
50
Klinger, D. A., & Brunson R. K. (2009). Police officers’ perceptual distortions during lethal
force situations: Informing the reasonableness standard. Criminology and Public Policy,
8, 117-140.
Lacks, J. R. (2008). The lone American dictatorship: How court doctrine and police culture limit
judicial oversight of the police use of deadly force. New York University Annual Survey
of American Law 64, 391-443.
Lee, C. (2018). Reforming the Law on Police Use of Deadly Force: De-Escalation, Preseizure
Conduct, and Imperfect Self-Defense. U. Ill. L. Rev., 629.
Lewinski, B. (2000). Why is the suspect shot in the back? The Police Marksman, 25, 20-28.
Lewinski, W. J., & Redmann, C. (2009). New developments in understanding the behavioral
science factors in the "stop shooting" response. Law Enforcement Executive Forum, 9,
35-54.
Lum, C. M., Stoltz, M., Koper, C. S., & Scherer, J. A. (2019). Research on Body-Worn Cameras:
What We Know, What We Need to Know. Criminology and Public Policy, 18(1), 1–29.
Mislevy, J. R. (1986). Recent developments in the factor analysis of categorical variables.
Journal of Educational Statistics, 11, 3-31.
Moon, B., & Zager, L. J. (2007). Police officers' attitudes toward citizen support: Focus no
individual, organizational and neighborhood characteristic factors. Policing - An
International Journal of Police Stratgies and Management, 30, 484-497.
Morin, R., Parker, K., Stepler, R., & Mercer, A. (2017). Behind the badge - Amid protests and
calls for reform, how police view their jobs, key issues and recent fatal encounters
between blacks and police. Retrieved from Pew Research Center website:
LEGAL AND COMMUNITY STANDARDS
51
http://assets.pewresearch.org/wp-content/uploads/sites/3/2017/01/06171402/Police-
Report_FINAL_web.pdf
Mourtgos, S. M., Mayer, R. C., Wise, R. A., & O-Rourke, H. (2019). The overlooked perspective
of police trust in the public: Measurement and effects on police job behaviors. Criminal
Justice Policy Review, 0887403419851850.
Murphy, D. W., & Worrall, J. L. (1999). Residency requirements and public perceptions of the
police in large municipalities. Policing: An International Journal of Police Strategies &
Management, 22, 327-342.
Muthen, B. O., & Hofacker, C. (1988). Testing the assumptions underlying tetrachoric
correlations. Psychometrika, 53, 563-578.
Nagin, D. S., & Telep, C. W. (2017). Response to “Procedural justice and policing: A rush to
judgment?”. Annual Review of Law and Social Science, 13, 55-58.
National Academies of Sciences, Engineering, and Medicine. (2018). Proactive policing: Effects
on crime and communities. Washington, DC: The National Academies Press.
Nieuwenhuys, A., Savelsbergh, G., & Oudejans, R. (2015). Persistence of threat-induced errors
in police officers' shooting decisions. Applied Ergonomics, 48, 263-272.
Nivette, A. E., & Eisner, M. (2013). Do legitimate polities have few homicides? A cross-national
analysis. Homicide Studies, 17, 3-26.
Owens, E., Weisburd, D., Amendola, K. L., & Alpert, G. P. (2018). Can you build a better cop?
Experimental evidence on supervision, training, and policing in the community.
Criminology & Public Policy, 17, 41-87.
Parker, K. D., Onyekwuluje, A. B., & Murty, K. S. (1995). African Americans' attitudes toward
the local police: A multivariate analysis. Journal of Black Studies, 25, 396-409.
LEGAL AND COMMUNITY STANDARDS
52
Pate, A. M., & Friddell, L. A. (1993). Police use of force: Official reports, citizen complaints,
and legal consequences. Washington, DC: Police Foundation.
Patil, S. V. (2018). Public support for the punishment of police use of force errors: Evidence of
ideological divergence and convergence. Police Quarterly, 21, 358-386.
Pickett, J., & Nix, J. (2019). Demeanor and police culture: Theorizing how civilian cooperation
influences police officers. Policing: An International Journal of Police Strategies and
Management.
President's Task Force on 21st Century Policing. (2015). Final report on the President's Task
Force on 21st Century Policing. Washington, DC: Office of Community Oriented
Policing Services.
Roche, S., & Roux, G. (2017). The "silver bullet" to good policing: a mirage - An analysis of the
effects of political ideology and ethnic identity on procedural justice. Policing: An
International Journal of Police Strategies and Management, 40, 514-528.
Sargeant, E., Murphy, K., & Madon, N. S. (2018). Is dissatisfaction with police inevitable?
Testing an integrated model of motivational postures and procedural justice in police-
citizen contacts. Police Practice and Research: An International Journal, 19(2), 125-137.
Schuck, A. M., & Rabe-Hemp, C. (2007). Women police: The use of force by and against female
officers. Women & Criminal Justice, 16, 91-117.
Scott v. Henrich, 39 F. 3d 912 (9th Cir. 1994).
Selby, N., Singleton, B., & Flosi, E. (2016). In context - Understanding police killings of
unarmed civilians. Florida: Contextual Press.
Shane, J., & Swenson, Zoe. (2019). Unarmed and dangerous: Patterns of threats by citizens
during deadly force encounters with police. New York, NY: Routledge.
LEGAL AND COMMUNITY STANDARDS
53
Shjarbak, J. A., Pyrooz, D. C., Wolfe, S. E., & Decker S. H. (2017). De-Policing and crime in the
wake of Ferguson: Racialized changes in the quantity and quality of policing among
Missouri police departments. Journal of Criminal Justice, 50, 42-52.
Silver, J. R., & Pickett, J. T. (2015). Toward a better understanding of politicized policing
attitudes: Conflicted conversvatism and support for police use of force. Criminology, 53,
650-676.
Small, S. (1983). Police and people in London: a group of young people (Vol. 2). London:
Policy Studies Institute.
Sparrow, M. K., Moore, M. H., & Kennedy, D. M. (1990). Beyond 911: A new era for policing.
New York: Basic Books.
Sunshine, J., & Tyler, T. R. (2003). The role of procedural justice and legitimacy in shaping
public support for policing. Law & Society Review, 37, 513-547.
Tankebe, J. (2013). Viewing things differently: The dimensions of public perceptions of police
legitimacy. Criminology, 51, 103-135.
Terrill, R. J. (1982). Complaint procedures: Variations on the theme of civilian participation.
Journal of Police Science and Adminstration, 10, 398-407.
Testa, P. & Deitrich, B. J. (2017). Seeing is believing: How video of police actions affects
criminal justice beliefs. Working Paper.
Thompson II, R. M. (2015). Police use of force: Rules, remedies, and reform. Washington D.C.:
Congressional Research Service.
Thompson, B. L., & Lee, J. D. (2004). Who cares if police become violent? Explaining approval
of police use of force using a national sample. Sociological Inquiry, 74, 381-410.
LEGAL AND COMMUNITY STANDARDS
54
Tiesman, H. M., Gwilliam, M., Konda, S., Rojek, J., & Marsh, S. (2018). Nonfatal injuries to law
enforcement officers: A rise in assaults. American Journal of Preventive Medicine, 54(4),
503–509.
Tuch, S. A., & Weitzer, R. (1997). The polls - trends: Racial differences in attitutes towards the
police. Public Opinion Quarterly, 61, 642-663.
Tyler, T. R. (1990). Why people obey the law: Procedural justice, legitimacy, and compliance.
Connecticut: Yale University Press.
Tyler, T. R., Boeckmann, R. J., Smith, H. J., & Huo, Y. J. (1997). Social justice in a diverse
society. Colorado: Westview Press.
Weber, M. (1946). Politics as a vocation. In Gerth, H., and Mills, C. W. (Eds.), From Max
Weber: Essays in Sociology. New York: Oxford University Press.
Weitzer, R., & Tuch, S. A. (1999). Race, class, and perceptions of discrimination by the police.
Crime and Delinquency, 45, 464-507.
Weitzer, R., & Tuch, S. A. (2002). Perceptions of racial profiling: race, class, and personal
experience. Criminology, 40, 435-456.
Wieber, A. (2017, December 5). 'We are still presented with situations that are unwinnable.' SLC
police explain use of deadly force in media demonstrations. The Salt Lake Tribune.
Retrieved from https://www.sltrib.com/news/2017/12/05/salt-lake-city-police-put-use-of-
force-tactics-on-display/
Wilson, G., & Dunham, R. (2001). Race, class, and attitudes towards crime control - The views
of the african american middle class. Criminal Justice and Behavior, 28, 259-278.
LEGAL AND COMMUNITY STANDARDS
55
Wolfe, S. E., Nix, J., Kaminski, R., & Rojek, J. (2016). Is the effect of procedural justice on
police legitimacy invariant? Testing the generality of procedural justice and competing
antecedents of legitimacy. Journal of Quantitative Criminology, 32, 253-282.
Worden, R. E., & McLean, S. J. (2014). Assessing police performance in citizen encounters:
Police legitimacy and management accountability. In Report to the National Institute of
Justice. John F. Finn Institute for Public Safety, Inc. Albany, NY.
Worrall, J. L. (1999). Public perceptions of police efficacy and image: the 'fuzziness' of support
for the police. American Journal of Criminal Justice, 24, 47-66.
Yang, K. (2005). Public administrators' trust in citizens: A missing link in citizen involvement
efforts. Public Administration Review, 65, 273-285.
LEGAL AND COMMUNITY STANDARDS
56
Appendix A
Table A1. Variable Operationalization and Descriptive Statistics
Variable
Mean
SD
Min
Max
Operationalization
Dependent variables
Use of Force
Perceptions
(Composite)
.671
.928
0
3
Count variable summed from “no”
answers to POLHITOK,
POLATTAK, and POLESCAP
POLHITOK
.301
.459
0
1
“Are there any situations you can
imagine in which you would
approve of a policeman striking an
adult male citizen?” 0 = Yes, 1 =
No
Years Asked: 1973, 75, 76, 78, 80,
83, 84, 86 - 2018
POLATTAK
.271
.444
0
1
“…Was attacking the policeman
with his fists?” 0 = Yes, 1 = No
Years Asked: 1973, 75, 76, 78, 80,
83, 84, 86 - 2018
POLESCAP
0.275
0.446
0
1
“…Was attempting to escape from
custody?” 0 = Yes, 1 = No
Years Asked: 1973, 75, 76, 78, 80,
83, 84, 86 - 2018
Independent variables
Year
2,003.74
7.72
1990
2018
Year of GSS survey
Age
46.52
17.28
18
89
Continuous measure of respondent
age
City Size
335.01
1161.48
0
0 - 8175
Ordinal measure of population of
respondent’s home city, in 1000s.
Income
29,891.28
31,860.04
0
$155,140
Ordinal measure of income, in
whole 1986 dollars.
Sex
1.55
.496
0
1
Reported gender; Male = 0,
female = 1.
Race
1.28
.592
1
3
Reported Race; white = 1, black
=2, other = 3.
Political Views
2.09
.787
1
3
Reported ideology; liberal = 1,
moderate =2, conservative =3.
Formal
Education
1.55
1.19
0
4
Respondent’s Highest Degree;
less than high school = 0, high
school = 1, junior college = 2,
bachelor degree = 3, graduate
degree = 4.
LEGAL AND COMMUNITY STANDARDS
57
Figure A1. Conditional ‘and’ Responses
Conditional “AND” Respondent Answers
General Social Survey, 1990-2018 (n=20,781)
69.94% 30.06%
53.27%
46.73%
15.78%
84.22%
83.74%
1.04%
98.96% 16.26% 93.35% 58.56% 41.44%
Ever Okay to Hit?
Escaping? Escaping?
Attacking Officer? Attacking Officer? Attacking Officer? Attacking Officer?
Would you ever approv e of a policeman…
1. …striking an adult male citizen?
2. …striking a citizen who was attempting
to escape from custody?
3. …striking a citizen who was attacking
the policeman with his fists?
“AN D ”
6.65%
