NUMBERS, FACTS AND TRENDS SHAPING THE WORLD

FOR RELEASE JAN. 11, 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
BY Rich Morin, Kim Parker, Renee Stepler and Andrew Mercer

FOR MEDIA OR OTHER INQUIRIES:
Kim Parker, Director of Social Trends Research
Rich Morin, Senior Editor
Molly Rohal, Communications Manager
202.419.4372
www.pewresearch.org

RECOMMENDED CITATION: Pew Research Center, Jan. 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.”

1
PEW RESEARCH CENTER

About Pew Research Center
Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes
and trends shaping America and the world. It does not take policy positions. The Center conducts
public opinion polling, demographic research, content analysis and other data-driven social
science research. It studies U.S. politics and policy; journalism and media; internet, science and
technology; religion and public life; Hispanic trends; global attitudes and trends; and U.S. social
and demographic trends. All of the Center’s reports are available at www.pewresearch.org. Pew
Research Center is a subsidiary of The Pew Charitable Trusts, its primary funder.
© Pew Research Center 2017

www.pewresearch.org

2
PEW RESEARCH CENTER

Table of Contents
About Pew Research Center

1

Terminology

3

Overview

4

1. Police culture

23

2. Inside America’s police departments

34

3. Police and the community

48

4. Police, fatal encounters and ensuing protests

60

5. Reimagining the police through training and reforms

68

6. Police views, public views

75

Acknowledgments

89

Methodology

90

www.pewresearch.org

3
PEW RESEARCH CENTER

Terminology
Throughout this report, “department” and “agency” are used interchangeably and refer to both
municipal police departments and county sheriff’s departments.
The terms “police officer” and “officer” refer to sworn officers in both police and sheriff’s
departments. References to “officers” or “all officers” include rank-and-file officers, sergeants and
administrators.
References to rank-and-file officers include sworn personnel assigned to patrol, detectives and
non-supervisory personnel assigned to specific units such as narcotics, traffic or community
policing. Sergeants are first-line supervisors. Administrators are officers with the rank of
lieutenant or higher, including senior command officers.
References to whites and blacks include only those who are non-Hispanic and identify as only one
race. Hispanics are of any race.
References to urban and suburban police officers are based on the ZIP code in which their
department is located. Urban police officers are defined as those whose department is within the
central city of a Metropolitan Statistical Area (MSA). Suburban officers are those whose
department is within an MSA, but not within a central city.

www.pewresearch.org

4
PEW RESEARCH CENTER

Behind the Badge
Amid protests and calls for reform, how
police view their jobs, key issues and recent
fatal encounters between blacks and police
Police work has always been hard. Today police say it is even
harder. In a new Pew Research Center national survey
conducted by the National Police Research Platform,
majorities of police officers say that recent high-profile fatal
encounters between black citizens and police officers have
made their jobs riskier, aggravated tensions between police
and blacks, and left many officers reluctant to fully carry out
some of their duties.
The wide-ranging survey, one of the largest ever conducted
with a nationally representative sample of police, draws on the
attitudes and experiences of nearly 8,000 policemen and
women from departments with at least 100 officers. 1 It comes
at a crisis point in America’s relationship with the men and
women who enforce its laws, precipitated by a series of deaths
of black Americans during encounters with the police that
have energized a vigorous national debate over police conduct
and methods.
Within America’s police and sheriff’s departments, the survey
finds that the ramifications of these deadly encounters have
been less visible than the public protests, but no less profound.
Three-quarters say the incidents have increased tensions
between police and blacks in their communities. About as
many (72%) say officers in their department are now less
willing to stop and question suspicious persons. Overall, more
than eight-in-ten (86%) say police work is harder today as a
result of these high-profile incidents.

1

See the methodology appendix for a more detailed description of the criteria used to select the agencies in the sample.

www.pewresearch.org

5
PEW RESEARCH CENTER

At the same time that black Americans are
dying in encounters with police, the number of
fatal attacks on officers has grown in recent
years. About nine-in-ten officers (93%) say
their colleagues worry more about their
personal safety – a level of concern recorded
even before a total of eight officers died in
separate ambush-style attacks in Dallas and
Baton Rouge last July.

Majority of police say fatal police-black
encounters are isolated incidents;
majority of the public says they point to
a bigger problem
% saying the deaths of blacks during encounters with
police in recent years are …
Isolated
incidents
Officers

67

Signs of a broader
problem
31

The survey also finds that officers remain
deeply skeptical of the protests that have
Public
39
60
followed deadly encounters between police
and black citizens. Two-thirds of officers
Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
(68%) say the demonstrations are motivated
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
to a great extent by anti-police bias; only 10%
PEW RESEARCH CENTER
in a separate question say protesters are
similarly motivated by a genuine desire to hold
police accountable for their actions. Some two-thirds characterize the fatal encounters that
prompted the demonstrations as isolated incidents and not signs of broader problems between
police and the black community – a view that stands in sharp contrast with the assessment of the
general public. In a separate Pew Research Center survey of U.S. adults, 60% say these incidents
are symptoms of a deeper problem.
A look inside the nation’s police departments reveals that most officers are satisfied with their
department as a place to work and remain strongly committed to making their agency successful.
Still, about half (53%) question whether their department’s disciplinary procedures are fair, and
seven-in-ten (72%) say that poorly performing officers are not held accountable.

Conflicting experiences and emotions mark police culture
Other survey findings underscore the duality of police work and the emotional toll that police work
can take on officers. About eight-in-ten (79%) say they have been thanked by someone for their
service in the month prior to the survey while on duty. But also during that time two-thirds say
they have been verbally abused by a member of their community, and a third have fought or
physically struggled with a suspect. A majority of officers (58%) say their work nearly always or
often makes them feel proud. But nearly the same share (51%) say the job often frustrates them.
More than half (56%) say their job has made them more callous.

www.pewresearch.org

6
PEW RESEARCH CENTER

Most police officers feel
respected by the public and, in
turn, believe officers have little
reason to distrust most people.
Rather than viewing the
neighborhoods where they
work as hostile territory,
seven-in-ten officers say that
some or most of the residents
of the areas they patrol share
their values. At the same time,
a narrow majority of officers
(56%) believe an aggressive
rather than courteous
approach is more effective in
certain neighborhoods, and
44% agree that some people
can only be brought to reason
the hard, physical way.

Police, public divided by race over whether attaining
equality requires more changes
% saying that …

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016; survey of U.S.
adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Long-standing tensions between police and blacks underlie many of the survey results. While
substantial majorities of officers say police have a good relationship with whites, Hispanics and
Asians in their communities, 56% say the same about police relations with blacks. This perception
varies dramatically by the race or ethnicity of the officer. Six-in-ten white and Hispanic officers
characterize police relations with blacks as excellent or good, a view shared by only 32% of their
black colleagues.
The racial divide looms equally large on other survey questions, particularly those that touch on
race. When considered together, the frequency and sheer size of the differences between the views
of black and white officers mark one of the singular findings of this survey. For example, only
about a quarter of all white officers (27%) but seven-in-ten of their black colleagues (69%) say the
protests that followed fatal encounters between police and black citizens have been motivated at
least to some extent by a genuine desire to hold police accountable.
And when the topic turns more broadly to the state of race relations, virtually all white officers
(92%) but only 29% of their black colleagues say that the country has made the changes needed to
assure equal rights for blacks. Not only do the views of white officers differ from those of their

www.pewresearch.org

7
PEW RESEARCH CENTER

black colleagues, but they stand far apart from those of whites overall: 57% of all white adults say
no more changes are needed, as measured in the Center’s survey of the general public.

Public, police differ on some key issues
Further differences in attitudes and perceptions emerge when the views of officers are compared
with those of the public on other questions. While two-thirds of all police officers say the deaths of
blacks at the hands of police are isolated incidents, only about four-in-ten members of the public
(39%) share this view while the majority (60%) believes these encounters point to a broader
problem between police and blacks.
And while a majority of Americans (64%) favor a ban on assault-style weapons, a similar share of
police officers (67%) say they would oppose such a ban.
On other issues the public and police broadly agree. Majorities of both groups favor the use of
body cameras by officers to record interactions with citizens (66% of officers and 93% of the
public). And about two-thirds of police (68%) and a larger share of the public (84%) believe the
country’s marijuana laws should be relaxed, and a larger share of the public than the police
support legalizing marijuana for both private and medical use (49% vs. 32%).
These findings come from two separate Pew Research Center surveys. The main survey is an
online poll of a nationally representative sample of 7,917 officers working in 54 police and sheriff’s
departments with 100 or more sworn officers. (Some 63% of all sworn officers work in
departments of this size.) The National Police Research Platform, headquartered at the University
of Illinois at Chicago during the study period, conducted this survey of police for the Pew Research
Center May 19-Aug. 14, 2016, using its panel of police agencies. The NPRP panel is described in
more detail in the methodology.
The views of the public included in this report drew from a Pew Research Center American Trends
Panel survey of 4,538 U.S. adults conducted online and by mail Aug. 16-Sept. 12, 2016. That survey
included many of the same questions asked on the police survey, allowing direct comparisons to be
made between the views of officers and the public.

Contrasting experiences, conflicting emotions
The survey provides a unique window into how police officers see their role in the community, how
they assess the dangers of the job and what they encounter on a day-to-day basis. It also gives a
glimpse into the psychology of policing and the way in which officers approach the moral and
ethical challenges of the job.

www.pewresearch.org

8
PEW RESEARCH CENTER

Police have a nuanced view of their role – they don’t see
themselves as just protectors or as enforcers. A majority (62%)
say they fill both of these roles equally. They also experience a
range of emotions on the job – often conflicting ones. A
majority of officers (58%) say their work nearly always or often
makes them feel proud. Almost as many (51%) say they nearly
always or often feel frustrated by the job.
Officers are somewhat less likely to say they feel fulfilled by
their job (42% say nearly always or often). Relatively few
officers (22%) say their job often makes them feel angry, but a
significant share (49%) say it sometimes makes them feel this
way. Officers who say their job often makes them feel angry
seem to be less connected to the citizens they serve. Fully 45%
say very few or none of the people in the neighborhoods they
serve share their values. Only 20% of officers who say they
hardly ever or never feel angry say the same.
White officers are significantly more likely than black officers to
associate negative emotions with their job. For example, 54% of
white officers say they nearly always or often feel frustrated by
their work, while roughly four-in-ten (41%) black officers say
the same. Hispanic officers fall in the middle on this measure. 2

Officers worry about their safety and think the public
doesn’t understand the risks they face
Fatal encounters between blacks and police have dominated the
headlines in recent years. But the story took on another twist
with the ambush-style attack that killed five police officers last
summer in Dallas. Because these attacks occurred while the
survey was in the field, it was possible to see if safety concerns
of officers were affected by the incidents by comparing views
before and after the assault.

Because of the small sample of non-Hispanic Asian officers (148), their views were not broken out separately but are included in the overall
results.

2

www.pewresearch.org

9
PEW RESEARCH CENTER

Overall, the vast majority of officers say they have serious concerns about their physical safety at
least sometimes when they are on the job. Some 42% say they nearly always or often have serious
concerns about their safety, and another 42% say they sometimes have these concerns. The share
of police saying they often or always have serious concerns about their own safety remained fairly
consistent in interviews conducted pre-Dallas to post-Dallas. 3
While physical confrontations are not a dayto-day occurrence for most police officers, they
are not altogether infrequent. A third of all
officers say that in the past month, they have
physically struggled or fought with a suspect
who was resisting arrest. Male officers are
more likely than their female counterparts to
report having had this type of encounter in the
past month – 35% of men vs. 22% of women.
And white officers (36%) are more likely than
black officers (20%) to say they have struggled
or fought with a suspect in the past month.
Among Hispanic officers, 33% say they had an
encounter like this.
Although police officers clearly recognize the
dangers inherent in their job, most believe the
public doesn’t understand the risks and
challenges they face. Only 14% say the public
understands these risks very or somewhat
well, while 86% say the public doesn’t
understand them too well or at all. 4 For their
part, the large majority of American adults
(83%) say they do understand the risks law
enforcement officers face.

White officers more likely than black
officers to have had a physical
altercation with a suspect
% of officers saying they have physically struggled or
fought with a suspect who was resisting arrest in the
past month
No

Yes

All officers

67

33

Men

65

35

Women

77

Whites
Blacks
Hispanics

22

64
80

36
20

67

33

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Police interactions with the public can range from casual encounters to moments of high stress.
And the reactions police report getting from community members reflect the diverse nature of

On July 7, 2016, five police officers were shot and killed, and nine officers were injured, in Dallas when they were ambushed by a black man
who claimed to be angry over recent police shootings of blacks. This incident occurred while the police survey was in the field; 6,957 officers
were interviewed before the Dallas shootings, and 960 were interviewed after the incident.
4 Throughout this report, whenever response options are combined, net shares are calculated before rounding.
3

www.pewresearch.org

10
PEW RESEARCH CENTER

those contacts. Large majorities of officers across most major demographic groups report that they
have been thanked for their service, but there are significant differences across key demographic
groups when it comes to verbal abuse. Men are more likely than women to say they have been
verbally abused by a community member in the past month. White and Hispanic officers are more
likely than black officers to have had this experience. And a much higher share of younger officers
(ages 18 to 44) report being verbally abused – 75%, compared with 58% of their older
counterparts.
The situations police face on the job can often present moral dilemmas. When asked how they
would advise a fellow officer in an instance where doing what is morally the right thing would
require breaking a department rule, a majority of police (57%) say they would advise their
colleague to do the morally right thing. Four-in-ten say they would advise the colleague to follow
the department rule. There’s a significant racial divide on this question: 63% of white officers say
they would advise doing the morally right thing, even if it meant breaking a department rule; only
43% of black officers say they would give the same advice.

www.pewresearch.org

11
PEW RESEARCH CENTER

Size and demographic composition of America’s police departments
According to the Bureau of Justice Statistics, there were 15,388 state and local law enforcement agencies,
employing more than 750,000 sworn officers in 2013 (the most recently available data). The majority of
these agencies were local municipal police departments (12,326). There were 3,012 sheriff’s
departments, which serve areas of the country that do not lie within the jurisdictions of police departments
of incorporated municipalities, though some small cities contract with the local sheriff’s department for
police services. There are also 50 primary state police agencies.
The Pew Research Center survey conducted by the National Police Research Platform is representative of
officers nationwide in local police and sheriff’s departments with at least 100 full-time police officers. A
majority of full-time sworn officers (477,317) were part of local police departments in 2013. Agencies with
at least 100 full-time equivalent officers employed 63% of the nation’s full-time officers even though these
departments accounted for only 5% of all local police departments (or 645 departments). It is these larger
local police agencies that have been at the center of recent national conversations on policing.

Majority of full-time officers are in
agencies with at least 100 officers
Number of
officers*
Total

Departments
Number Percent

Full-time officers
Number

Percent

12,326

100

477,317

100

1,000 or more

49

<0.5

161,883

34

596

5

138,248

29

10-99

5,786

47

156,262

33

1-9

5,895

48

20,926

4

100-999

*Includes both full-time and part-time officers with a weight of 0.5
for part-time employees.
Source: Bureau of Justice Statistics, Law Enforcement Management
and Administrative Statistics Survey (LEMAS), 2013.
“Behind the Badge”
PEW RESEARCH CENTER

In 2013, more than a quarter (27%) of full-time police officers were racial or ethnic minorities. Some 12%
of full-time local police officers were black; another 12% were Hispanic; and 4% were Asian, Pacific
Islander, American Indian or multiracial. By comparison, 12% of the U.S. adult population was black; 15%
was Hispanic; and 8% was some other racial or ethnic minority in 2013. Women remain underrepresented
in the field, making up just 12% of full-time officers in 2013; women are 51% of the U.S. adult population.
In general, departments that serve larger populations are more racially and ethnically diverse and tend to
have a higher share of women serving as full-time officers. For example, more than four-in-ten officers
serving populations of 500,000 or more were racial or ethnic minorities in 2013, compared with fewer
than one-in-five in jurisdictions with less than 50,000 people.

www.pewresearch.org

12
PEW RESEARCH CENTER

Police are highly committed to their work but say more officers are
needed
A look inside the nation’s police departments reveals a great deal about how officers view their
jobs, their leadership and their resources. For the most part police officers give their workplace a
positive rating and are committed to their agency’s success. A solid majority of officers are either
very satisfied (16%) or satisfied (58%) with their agency as a place to work. And an overwhelming
share of officers (96%) agree that they are strongly committed to making their agency successful.
Still, police do not offer universal praise of their departmental leadership. Only three-in-ten say
they are extremely (7%) or very (23%) supportive of the direction that top management is taking
their organization. About half are moderately (28%) or slightly (19%) supportive and 15% are not
supportive at all.
And police express serious concerns about resource limitations. At the most basic level, most
police (86%) say their department does not have enough officers to adequately police the
community. Police who work in larger agencies (with 1,000 officers of more) are more likely than
those working in smaller agencies to say that there is a shortage of officers in their department
(95% vs. 79%).
Police give their departments relatively
positive, though not exemplary, ratings for
training and equipping officers to do their
jobs. Roughly four-in-ten officers (39%) say
their department has done very well in terms
of training them adequately for their job, and a
similar share (37%) give their department high
marks for clearly communicating the
responsibilities of the job. About three-in-ten
officers (31%) say their department has done
very well when it comes to equipping them to
perform their job. On each of these
dimensions, about four-in-ten officers say
their department has done somewhat well,
while about one-in-five rate their department’s
performance as not too well or not at all well
in these areas.

Most officers say their department has
too few officers to police the community
% of officers saying their department ___ have enough
officers to adequately police the community
Does not
All officers

Does

86

13

Agency size
<1,000 officers
1,000 or more 95

79

21
5

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

13
PEW RESEARCH CENTER

Again there are gaps by department size, with smaller departments (1,000 officers or fewer) giving
their leadership significantly higher ratings when it comes to training and equipping them, as well
as communicating job expectations.

Most officers say their use-of-force guidelines are appropriate and helpful
As many departments grapple with use-of-force policies and training, most officers say their own
agency’s guidelines strike the right balance. About one-in-four (26%) say the rules governing use
of force in their department are too restrictive, while 73% say they are about right (1% say the
guidelines are not restrictive enough).
Roughly a third (34%) of officers say their department’s guidelines are very useful when police are
confronted with actual situations where force may be necessary. An additional 51% say the
guidelines are somewhat useful. Some 14% say they are not too useful or not at all useful. And
when the department guidelines are not being followed, police overwhelmingly say fellow officers
need to step up. Fully 84% say officers should be required to intervene when they believe another
officer is about to use unnecessary force; just 15% say they should not be required to intervene.
In terms of striking the right balance between acting decisively versus taking time to assess a
situation, police tend to be more concerned that officers in their department will spend too much
time diagnosing a situation before acting (56% worry more about this) than they are about officers
not spending enough time before acting
decisively (41%).
About half say disciplinary process in
Black officers are much more likely than white
or Hispanic officers to say they worry more
that officers will not spend enough time
diagnosing a situation before acting (61% for
blacks vs. 37% for whites and 44% of
Hispanics). Overall, blacks and department
administrators (59%) are the only two major
groups in which a majority is more concerned
that officers will act too quickly than worry
that they will wait too long before responding
to a situation.
Officers give their departments mixed ratings
for their disciplinary processes. About half

their department is fair

% of officers saying they ___ with each of the following
statements
Disagree
The disciplinary process
in their agency is fair
Officers who consistently
do a poor job are held
accountable

Agree

53

72

45

27

Note: “Agree” comprises those who say they agree or strongly agree
with the statement. “Disagree” comprises those who say they
disagree or strongly disagree with the statement. No answer
category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

14
PEW RESEARCH CENTER

(45%) agree that the disciplinary process in their agency is fair, while 53% disagree (including onein-five who strongly disagree). When they are asked more specifically about the extent to which
underperforming officers are held accountable, police give more negative assessments of their
departments. Only 27% agree that officers who consistently do a poor job are held accountable,
while 72% disagree with this.

Most officers have had at least some training in key areas of reform
Reforming law enforcement tactics and procedures – particularly as they relate to the use of force
– has become an important focus both inside and outside the police department. In the wake of
recent fatalities of blacks during encounters with police, recommendations have been made to
prevent these types of situations from occurring.
The survey finds broad support among police,
especially administrators, for the use of body
cameras. Even so, officers are somewhat
skeptical that their use would change police
behavior. Half of all officers say body cameras
would make police more likely to act
appropriately, while 44% say this wouldn’t
make any difference.
Despite the national attention given to training
and reforms aimed at preventing the use of
unnecessary force, relatively few (half or fewer
rank-and-file officers) report having had at
least four hours of training in some specific
areas over the preceding 12 months.

Roughly two-thirds of officers say they
favor the use of body cameras
% of officers saying they personally ___ the use of body
cameras, regardless of whether their department uses
them
Oppose

Favor

All officers

33

66

Rank-and-file
officers

34

65

Sergeants
Administrators

31

67
18

82

About half of rank-and-file officers say they
Note: No answer category not shown.
have had at least four hours of firearms
Source: Survey of law enforcement officers conducted May 19-Aug.
training in the last 12 months involving shoot14, 2016.
“Behind the Badge”
don’t shoot scenarios (53%) and nonlethal
PEW RESEARCH CENTER
methods to control a combative or threatening
individual (50%). Some 46% of officers have
had at least four hours of training in how to deal with individuals who are having a mental health
crisis, and 44% say they have had at least four hours of training in how to de-escalate a situation so
it is not necessary to use force.

www.pewresearch.org

15
PEW RESEARCH CENTER

About four-in-ten officers say they have received at least four hours of training in bias and fairness
(39%) and how to deal with people so they feel
Impact of fatal incidents involving
they’ve been treated fairly and respectfully
blacks felt more by large departments
(37%).

than by small agencies

Most officers say high-profile
incidents have made policing
harder

% of officers in ___ saying each has happened in their
department as a result of high-profile incidents
involving blacks and the police

Whether an officer works in a department that
employs hundreds or thousands of sworn
officers or is located in a quiet suburb or
bustling metropolis, police say their jobs are
harder now as a consequence of recent highprofile fatal incidents involving blacks and
police.

Small departments (fewer than 300 officers)
Large departments (2,600 or more)
Officers have become less willing to stop and
question people who seem suspicious
54
86
Interactions between police and blacks have become
more tense
61

Overall, fully 86% of officers say their jobs are
harder, including substantial majorities of
officers in police departments with fewer than
300 officers as well as those working in “mega
departments” with 2,600 officers or more (84%
and 89%, respectively). In fact, across every
major demographic group analyzed for this
survey, about eight-in-ten officers or more say
these high-profile incidents have made policing
more challenging and more dangerous.
While the impact of these incidents is broadly
felt, officers in larger departments are far more
likely than those in small agencies to say these
incidents have had an impact. For example,
roughly half of officers (54%) in departments
with fewer than 300 officers say their peers
have become less willing to stop and question
people who seem suspicious. By contrast, fully
86% of police in departments with 2,600
officers or more say fellow officers are now

87
Officers have become more reluctant to use force
when it is appropriate
63
85
Officers have become more concerned about their
safety
88
95
Departmental changes
The department has modified its policies or procedures
about the use of force
19
68
The department has taken steps to improve relations
between police and blacks
35
66
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

16
PEW RESEARCH CENTER

more hesitant to question people who look or act suspicious. Similarly, roughly nine-in-ten officers
(87%) in the largest departments say that police interactions with blacks have become more tense;
61% of officers in small departments agree.
Police in larger departments also are more likely than those in small agencies to say officers in
their department are more reluctant to use force to control a suspect even when it is appropriate, a
move that police critics may view as a positive sign but others may see as putting officers at
increased risk.
The survey also found that roughly half (46%) of officers say fatal encounters between blacks and
police in recent years have prompted their department to modify their use-of-force policies.
Officers in large departments are more than three times as likely to report that their departments
have made this change as small agencies (68% vs. 19%). About two-thirds of police in larger
departments (66%) say their departments have taken steps to improve relations with black
residents. By contrast, about a third of officers in small departments (35%) have made similar
outreach efforts.

How officers view police relations
with whites and minorities
Large majorities of white, black and Hispanic
officers agree that police and whites in their
communities get along. 5
But striking differences emerge when the focus
shifts to police relations with racial and ethnic
minorities. A consistently smaller share of
black officers than their white or Hispanic
colleagues say the police have a positive
relationship with minorities in the community
they serve. Roughly a third of all black officers
(32%) characterize relations with blacks in
their community as either excellent or good,
while majorities of white and Hispanic officers
(60% for both) offer a positive assessment.

Most white, Latino officers say fatal
encounters between blacks and police
are isolated incidents; majority of black
officers disagree
% of officers saying the deaths of blacks during
encounters with police in recent years are …
Isolated
incidents
All officers

Signs of a
broader problem

67

31
0

Whites

72

Blacks
Hispanics

27
43

72

57
26

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

The sample includes 148 non-Hispanic Asian officers, too small to reliably characterize Asian officers as a group or compare them with other
racial and ethnic groups. The views of Asian officers, however, are included in the overall results.

5

www.pewresearch.org

17
PEW RESEARCH CENTER

At the same time, only about half of all black officers (46%) but large majorities of Hispanic (71%)
and white (76%) officers say relations between police and Hispanics are excellent or good.
Similarly, three-quarters of all black officers but 91% of white officers and 88% of Hispanic officers
rate relations with Asians in their communities positively.

Majority of police view fatal encounters as isolated incidents
Two-thirds of police officers (67%) say the highly publicized fatal encounters between police and
blacks are isolated incidents, while 31% describe them as signs of a broader problem. Yet
underlying this result are striking differences between the views of black and white officers –
differences that mirror the broader fault lines in society at large on racial issues.
A majority of black officers (57%) say these encounters are evidence of a broader problem between
police and blacks, a view held by only about a quarter of all white (27%) and Hispanic (26%)
officers.
Black female officers in particular are more
likely to say these incidents signal a more farreaching concern. Among all sworn officers,
63% of black women say this, compared with
54% of black men.

Most officers say protests mainly
motivated by bias toward police
% of officers saying protests over deaths of blacks who
died during encounters with the police are motivated __
by …
A great deal

Some

Not much

Not at all

Long-standing bias against the police

Widespread doubts about protesters’
motives

NET 92%

Most police officers are deeply skeptical of the
motives of the demonstrators who protested
after many of the deadly encounters between
police and blacks. Fully nine-in-ten (92%)
believe that long-standing bias against the
police is a great deal (68%) or some (24%) of
the motivation behind these demonstrations.
In sharp contrast, only about a third (35%) of
officers say in a separate question that a
genuine desire to hold officers accountable for
their actions is at least some of the motivation
for the protests.

68

24

42

A genuine desire to hold officers accountable for their
actions
NET 35%
10

25

36

28

Note: No answer category not shown. NETs calculated before
rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

18
PEW RESEARCH CENTER

Once again, race pushes police in opposite directions. Among black officers, 69% say the protests
were sincere efforts to force police accountability – more than double the proportion of whites
(27%) who share this view. Female officers, older police and department administrators also are
more likely than male officers, younger police and rank-and-file officers to believe protesters
genuinely seek police accountability.

Support for aggressive, physical tactics
The law gives police great discretion in how they interact with citizens. Depending on the
situation, these techniques can range from polite persuasion to the use of forceful and more
pointed verbal commands to the extreme physical measures that officers sometimes use, often as a
last resort, to control threatening or combative individuals. The use of these more severe
techniques has been a main
focus of the national debate
over police methods.
To measure their attitudes
toward more aggressive
tactics, officers were asked
how much they agreed or
disagreed with two
statements. The first
statement read, “In certain
areas of the city it is more
useful for an officer to be
aggressive than to be
courteous.” The second
measured support for the
assertion that “some people
can only be brought to reason
the hard, physical way.”

Some officers say tough, aggressive tactics are
needed with some people and in some neighborhoods
% of officers saying they __ with each of the following statements
Strongly disagree

Disagree

Agree

Strongly agree
NET 44%

Some people can only be
brought to reason the hard,
physical way
In certain areas of the city it
is more useful for an officer
to be aggressive than to be
courteous

10

45

39

5

NET 56%
10

34

39

17

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

A narrow majority of officers (56%) feel that in some neighborhoods being aggressive is more
effective than being courteous, while 44% agree or strongly agree that hard, physical tactics are
necessary to deal with some people.

www.pewresearch.org

19
PEW RESEARCH CENTER

On both measures, a larger share of younger, less senior officers and those with less than five years
of experience favor these techniques, while proportionally fewer older, more experienced officers
or department administrators endorse them.

A majority of officers say they have become more callous
There’s a saying in police work that officers see things the public doesn’t see – and also things the
public shouldn’t see. Exposure to the dark side of life, coupled with the stress that officers
encounter working in high-pressure situations or with hostile individuals, means that many
officers may pay an emotional price for their service.
For example, a 56% majority of officers say they have become more callous toward people since
taking their job. Younger officers and white officers are more likely than older or black officers to
say they have become more callous.
Officers who report they have grown more callous are also more likely than their colleagues to
endorse aggressive or physically harsh tactics with some people or in some parts of the
community. They also are more likely than other officers to say they are frequently angered or
frustrated by their jobs or to have been involved in a physical or verbal confrontation with a citizen
in the past month or to have fired their service weapon while on duty at some point in their
careers.
It is difficult to discern with these data whether increased callousness is a primary cause or a
consequence of feelings of anger or frustration, or attitudes toward aggressive tactics. However,
the data suggest that these feelings and behaviors are related. For example, officers who sense they
have become more callous on the job are about twice as likely as those who say they have not to say
their job nearly always or often makes them feel angry (30% vs. 12%). They also are more likely to
feel frustrated by their job (63% vs. 37%).
Among those officers who say they have become more callous, about four-in-ten (38%) physically
struggled or fought with a suspect in the previous month compared with 26% of those who say
they have not become more insensitive.

www.pewresearch.org

20
PEW RESEARCH CENTER

Similarities and differences between police and public views
On a range of issues and attitudes, police and the public often see the world in very different ways.
For example, when both groups are asked whether the public understands the risks and rewards of
police work, fully eight-in-ten (83%) of the
public say they do, while 86% of police say they
don’t – the single largest disparity measured in
these surveys.
And while the country is divided virtually down
the middle over the need to continue making
changes to obtain equal rights for blacks, the
overwhelming majority of police (80%) say no
further changes are necessary. The public also is
twice as likely as police to favor a ban on assaultstyle weapons (64% vs. 32%).
Yet these differences in views are matched by
equally significant areas of broad agreement.
Large majorities of officers (92%) and the public
(79%) say anti-police bias is at least somewhat of
a motivation for those protesting the deaths of
blacks at the hands of police. Majorities of police
and the public favor the use of body cameras by
officers, though a significantly larger share of the
public supports their use (93% vs. 66%) and sees
more benefits from body cams than the police
do.
While they disagree about an assault weapons
ban, large majorities of the police (88%) and the
public (86%) favor making private gun sales and
sales at gun shows subject to background checks.
Majorities also favor creating a federal database
to track all gun sales (61% for police and 71% of
the public).

www.pewresearch.org

21
PEW RESEARCH CENTER

The remainder of this report explores in greater detail the working lives, experiences and attitudes
of America’s police officers. Chapter 1 examines police culture, how officers view their job as well
as the risks and rewards of police work. Chapter 2 reports how officers view their departments and
their superiors as well as officers’ attitudes toward the internal rules and policies that govern how
they do their job, including the use of force. Chapter 3 looks at how officers view the citizens they
serve and how they think the citizens view them, including officers’ perceptions of the relations
between police and whites, blacks and other minority groups in their communities. Chapter 4
explores police reaction to recent fatal encounters between blacks and police, the protests that
followed many of these incidents and the impact those events have had on how officers do their
job. Chapter 5 looks at how officers view various police reforms, including the use of body
cameras, and reports on the kinds of police training officers receive to help reduce bias, deescalate threatening situations as well as how to know when – and when not to – use their service
weapons or use deadly force. The final chapter compares and contrasts the views of police with
those of the public on a wide range of issues relevant to police work, including attitudes toward
gun law reforms and changes to the country’s marijuana laws. It also explores how each group
views recent fatal encounters between blacks and police as well as the protests that have frequently
followed those incidents.
Other key findings:
•

About half of black officers (53%) say that whites are treated better than minorities in their
department or agency when it comes to assignments and promotions. Few Hispanic (19%)
or white officers (1%) agree. About six-in-ten white and Hispanic officers say minorities
and whites are treated the same (compared with 39% of black officers).

•

Most officers say that outside of required training, they have not discharged their service
firearm while on duty; 27% say they have done this. Male officers are about three times as
likely as female officers to say they have fired their weapon while on duty – 30% of men vs.
11% of women.

•

Roughly three-in-ten officers (31%) say they have patrolled on foot continuously for 30
minutes or more in the past month; 68% say they have not done this.

•

Officers are divided over whether local police should take an active role (52%) in
identifying undocumented immigrants rather than leaving this task mainly to federal
authorities (46%).

www.pewresearch.org

22
PEW RESEARCH CENTER

•

The share of sworn officers who are women or minorities has increased slowly in recent
decades. Since 1987 the share of female officers has grown from 8% to 12% in 2013, the last
year the federal Department of Justice measured the demographic characteristics of police
agencies. During that time, the share of black officers increased from 9% to 12% while the
Hispanic share more than doubled, from 5% to 12%.

•

About seven-in-ten officers say some or most of the people in the neighborhoods where
they routinely work share their values and beliefs. Officers in larger departments are less
likely than those in smaller departments to say they share values with the people in the
areas where they patrol.

•

About half (51%) of police officers compared with 29% of all employed adults say their job
nearly always or often frustrates them, while about four-in-ten officers (42%) and half of
employed adults (52%) say their work frequently makes them feel fulfilled.

•

A large majority of all officers (76%) say that responding effectively to people who are
having a mental health crisis is an important role for police. An additional 12% say this is a
role for them, though not an important one and 11% say this is not a role for police.

www.pewresearch.org

23
PEW RESEARCH CENTER

1. Police culture
In police departments around the country, officers are often sent out to start their shift with an
admonition from their supervisor – something to the effect of, “Come back safe.” There’s no
disputing that law enforcement is a dangerous occupation. 6 Police officers feel this acutely, but
they question whether the public truly understands the risks they face on the job. The vast
majority of police (84%) say they worry about their safety at least some of the time, and roughly
the same share (86%) say they don’t think the public understands the risks and challenges they
face on the job.
While physical confrontations are not a part of the daily routine for most police officers, a third of
all officers say they have struggled or fought with a suspect who was resisting arrest in the past
month. Some 27% say they have discharged their service firearm while on duty at some point in
their career, not including anytime they used their weapon in training exercises.
The survey captured the duality of police work on several dimensions. A majority of police (58%)
say their work as a law enforcement officer
nearly always or often makes them feel proud.
About six-in-ten police officers see
But nearly the same share (51%) say their work
themselves as protectors and enforcers
often makes them feel frustrated. A large
% of officers saying, even if both are important parts of
majority (79%) say they have been thanked by
their work, they see themselves more as …
someone for their police service in the past
month, but almost as many (67%) say they
Both equally
Protector
Enforcer
have been verbally abused by a member of
All officers
62
31
8
their community while on duty during that
same period. And when asked whether they
view themselves more as protectors or
Rank-and-file
R&F officers
62
29
8
enforcers, roughly six-in-ten police officers
officers
(62%) say they fill both of these roles equally.
There are significant gaps across key
demographic groups on several of these
measures. White officers are more likely than
black officers to believe that the public doesn’t
understand the risks they face on the job, to
say they have recently gotten into a physical

Sergeants

59

35

Administrators

57

41

6

2

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

According to the Bureau of Labor Statistics, police officers are at a higher risk of suffering work-related injuries or fatalities than workers in
most other occupations. See http://www.bls.gov/iif/oshwc/cfoi/police-officers-2014.htm for more details.

6

www.pewresearch.org

24
PEW RESEARCH CENTER

struggle with a suspect and to say the job makes them feel frustrated or angry. Male and female
officers have similar outlooks on their job but report different experiences, especially when it
comes to violent confrontations.
Experiences and attitudes also differ substantially according to rank. For example, compared with
administrators, rank-and-file officers and sergeants worry more about their safety and feel less
understood by the public.

While officers worry about their safety, most feel public doesn’t
understand the risks they face
When asked at the most basic level whether they see themselves more as protectors or enforcers,
most police officers (62%) say they see themselves filling both roles equally. About three-in-ten
(31%) say they see themselves mainly as a protector, while only 8% say they view themselves more
as enforcers.
Across gender, racial and age lines, majorities
of police officers say they consider themselves
both protectors and enforcers. Black officers
(69%) are more likely than white officers
(59%) to say this, while a somewhat higher
share of white officers (32%) than black
officers (27%) say they see themselves mainly
as protectors. Administrators are somewhat
more likely than sergeants or rank-and-file
officers to view themselves mainly as
protectors.
Whether protectors or enforcers, most police
officers say that they face dangers on the job.
Roughly four-in-ten say that they nearly
always (13%) or often (29%) have serious
concerns about their physical safety when they
are at work. Another 42% say they sometimes
have these concerns. Only 16% say they hardly
ever or never worry about their safety.

Younger officers worry more often about
their physical safety
% of officers saying they have serious concerns about
their physical safety when they are at work …
Nearly always/
Often
All officers

Sometimes

42

Ages 18-44

42

16

41

46

45+

Hardly ever/
Never

43

37

13
20

Rank-and-file
R&F officers
officers

44

41

15

Sergeants

44

43

14

Administrators

26

47

26

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

25
PEW RESEARCH CENTER

Male and female officers report having serious concerns about their safety with roughly equal
frequency. And there are only modest differences on this measure between black and white
officers.
Younger officers express a higher level of concern about their safety than do their older
counterparts. Among those ages 18 to 44, fully 46% say they nearly always or often have serious
concerns about their physical safety when they are at work; 37% of officers ages 45 and older voice
the same level of concern.
Perhaps related to the age differences or their
specific job assignments, rank-and-file officers
and sergeants are more likely than
administrators to say they often have concerns
about their safety.
The level of concern on the part of individual
officers did not appear to increase after the
fatal attack on five officers in Dallas last
summer. The share of police saying they nearly
always or often have serious concerns about
their own safety remained fairly consistent
pre-Dallas to post-Dallas.

While not commonplace, physical
confrontations with suspects are not a rare
occurrence for police
A third of all officers say they have physically
struggled or fought in the past month with a
suspect who was resisting arrest. There are
large demographic differences on this
measure. Male officers are much more likely
than their female counterparts to report
having had this type of encounter in the past
month – 35% of men vs. 22% of women.
There are large gaps by race as well. While
36% of white officers say they have struggled

Men, newer officers more likely to
report violent encounters with resistant
suspects
% of officers saying, in the past month, they have
physically struggled or fought with a suspect who was
resisting arrest
No

Yes

All officers

67

33

Men

65

35

Women

77

Whites
Blacks

22

64

36

80

Hispanics

20
67

Rank-and-file
R+F officers
officers
Sergeants

33

64

36

71

28

Administrators 88

11

Years in law enforcement
Less than 5
5 or more

47
70

53
30

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

26
PEW RESEARCH CENTER

or fought with a suspect in the past month, only 20% of black officers say they have had the same
type of experience. Among Hispanic officers, 33% say they had an encounter like this. These racial
patterns persist across different-sized agencies.
Officers who are fairly new to the job are much more likely than more seasoned officers to have
struggled or fought with a suspect in the past month. Some 53% of officers who have been on the
job for less than five years say they have done this, compared with 30% of those who have been on
the job five years or longer.
These experiences also vary significantly by rank. Among rank-and-file officers, 36% say they have
had a violent encounter with a suspect who was resisting arrest in the past month; 28% of
sergeants report having a run-in like this, as do 11% of administrators.
Perhaps not surprisingly, police who have had this type of experience recently are more likely than
those who have not to say they often have serious concerns about their personal safety when they
are on the job (54% vs. 37%).

About seven-in-ten officers have never fired
their service weapon while on duty
Most officers say that, outside of required
training, they have not discharged their service
firearm while on duty; 27% say they have done
this. Male officers are about three times as
likely as female officers to say they have fired
their weapon while on duty – 30% of men vs.
11% of women.
Among white officers, 31% say they have
discharged their service firearm while on duty.
A smaller share of black (21%) and Hispanic
(20%) officers report doing the same.
Not surprisingly, officers with a longer tenure
in law enforcement are more likely than newer
officers to have used their firearm while on
duty. Some 14% of those with less than five
years of experience say they have fired their

Male officers, whites more likely to have
fired service weapon on duty
% of officers saying, other than on a gun range or while
training, they have discharged their service firearm
while on duty
No
All officers

72

Men

70

Women

88

Whites

Yes
27

30
11

69

31

Blacks

79

21

Hispanics

79

20

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

27
PEW RESEARCH CENTER

weapon, while 29% of officers who have been on the job five years or longer say they have done
this. Interestingly, those with 20 or more years of experience are no more likely than those with 10
to 19 years of service to have used their weapon while on duty.

How well does the public understand the risks police face? Four-in-ten officers say not well
at all
The relationship between the police and the
public is a complicated one. While many police
officers often worry about their own safety, the
vast majority believe the public doesn’t
understand the risks and challenges they face
on the job. When asked how well the public
understands the risks police face, 46% of
officers say not too well, and an additional
40% say not well at all. Only 1% say the public
understands the risks and challenges faced by
police very well, and 12% say the public
understands somewhat well.
White and Hispanic officers are much more
likely than black officers to say the public
doesn’t understand the risks and challenges
involved in police work. Fully 42% of white
and Hispanic officers say the public does not
understand these risks well at all, compared
with only 29% of black officers.

Most police say the public doesn’t
understand the risks they face
% of officers saying the public understands the risks and
challenges they face on the job …
Very well/
Somewhat well
All officers

14

Not too well
46

Not well at all
40

% saying not well at all, by subgroup
Whites
Blacks

42
29

Hispanics

42

Ages 18-44

44

45+

34

<1,000 officers

35

Agency size
1,000 or more

45

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race. “Very well/Somewhat
well” calculated before rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”

Younger officers are more skeptical than their
PEW RESEARCH CENTER
older counterparts that the public grasps the
risks and challenges police face on the job.
While 44% of officers ages 18 to 44 say the public doesn’t understand these things well at all, only
34% of officers ages 45 and older agree with this assessment. Similarly, rank-and-file officers
(42%) and sergeants (39%) are more likely than administrators (25%) to say the public doesn’t
understand police work well at all.

Officers from larger departments feel more of a disconnect with the public than officers from
smaller departments. Among those in departments with 1,000 or more officers, 45% say the public

www.pewresearch.org

28
PEW RESEARCH CENTER

doesn’t understand well at all the risks and challenges police face. Among those in departments
with fewer than 1,000 officers, 35% share this view.

For police, contact with citizens can be a mixed bag
Police interact with the public in a variety of
settings, and the reaction they get from
community members can range from
affirmative to abusive. About eight-in-ten
officers (79%) say they have been thanked by a
community member for their police service in
the past month. At the same time, 67% say
they have been verbally abused by a member
of their community while they were on duty
over the same period.
In some ways, these two common experiences
sum up the complex nature of policing: Praise
and hostility can both be part of a typical day’s
work. Fully 55% of officers say they have had
both of these experiences over the past month.

For many officers, being thanked and
verbally abused are commonplace
% of officers saying they have been ___ by a community
member in the past month while on duty
Thanked for their
service

79

Verbally abused

Had both of these
experiences

67

55

Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

The experience of being thanked by a community member tends to be fairly universal. Large
majorities of officers across most major demographic groups report that they have been thanked
for their service in the past month. The share of Hispanic officers reporting this (73%) is somewhat
lower than the share of white (81%) or black (83%) officers.
Officers who are relatively new to the job are more likely to report being thanked in the past
month: 92% of officers who have been on the job less than five years say this, compared with 77%
of those who have been on the force five years or longer.
When it comes to being verbally abused by community members while on duty, male officers
report having had this experience more often than their female counterparts: 69% of men say
they’ve had this type of experience in the past month, compared with 60% of women.
The gap between black officers and their white and Hispanic counterparts is even wider: Seven-inten white officers and roughly the same share of Hispanic officers (69%) say they have been
verbally abused by a community member in the past month, while 53% of black officers report the

www.pewresearch.org

29
PEW RESEARCH CENTER

same. In addition, younger officers are more
likely than older officers to have had this type
of encounter – 75% among officers ages 18 to
44 vs. 58% among those ages 45 and older.
Administrators, who have less day-to-day
contact with community members, are
significantly less likely to say they have been
verbally abused in the past month. Even so,
43% of administrators say they have had this
type of experience.

Roughly three-in-ten officers say they’ve
spent at least 30 consecutive minutes
patrolling on foot in the past month
The image of the “cop walking the beat” is not
necessarily the norm for today’s police. Among
all officers, regardless of rank or assignment,
roughly three-in-ten officers (31%) say they
have patrolled on foot continuously for 30
minutes or more in the past month; 68% say
they have not done this.
Among officers who say their current job
responsibility is patrolling (45% of all police),
39% say they have patrolled on foot for 30
minutes or more in the past month. Within
this group, similar shares of men and women
say they have patrolled on foot in the past
month. And younger and older patrol officers
are about equally likely to have done this.

White and Hispanic officers, more than
blacks, report being verbally abused by
community members
% of officers saying they have been verbally abused by a
member of the community in the past month while on
duty
No

Yes

All officers

32

67

Men

31

69

Women

39

Whites
Blacks

30
47

Hispanics

70
53

31

Ages 18-44

69

25

45+

75

42

Rank-and-file
R&F officers
officers

58

30

Sergeants
Administrators

60

70

33
56

66
43

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

The prevalence of patrolling on foot doesn’t differ significantly by community type – similar shares
of patrol officers from urban (39%) and suburban (38%) departments say they have patrolled on
foot for at least 30 continuous minutes in the past month.
To be sure, there are many ways – beyond walking the streets – for officers to engage with their
communities. A third of all officers say they have spoken to a citizens’ group or a school or

www.pewresearch.org

30
PEW RESEARCH CENTER

appeared at a community event in the past
month. Where administrators are less likely
than officers to patrol the streets, they are
about twice as likely to appear at community
events. Among rank-and-file officers, 31% say
they have done this in the past month. By
comparison, 59% of administrators (and 37%
of sergeants) have appeared at a community
event.
Police from larger agencies (1,000 officers or
more) are less likely than those from agencies
with fewer than 1,000 officers to have spoken
to a citizens’ group or participated in a
community event (28% vs. 38%). Black
officers are somewhat more likely than white
officers to have done this (42% vs. 32%).

Police say they feel pride in their
work more often than fulfillment
Police express a range of emotions when asked
how their work makes them feel. Pride is a
common sentiment, but so is frustration. A
majority of all officers say their work in law
enforcement nearly always (23%) or often
(35%) makes them feel proud. About half say
their work nearly always (10%) or often (41%)
makes them feel frustrated.

Administrators about twice as likely as
rank-and-file officers to have
participated in a community event in the
past month
% of officers saying, in the past month, they have spoken
to a citizens’ group or a school or appeared at a
community event

All officers

66

33

Rank-and-file officers

69

31

Sergeants

37

63

Administrators

59

40

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

About six-in-ten officers say their work
nearly always or often makes them feel
proud
% of officers saying their work as a law enforcement
officer ___ makes them feel …
Nearly always/
Often
Proud

To some degree, pride and frustration are
negatively correlated. Among the small share
of officers who say their work as a police
officer hardly ever or never makes them feel
proud, the vast majority (90%) say they nearly
always or often feel frustrated by their work.
Among those who nearly always or often feel

Yes

No

7
16

42

42
22

9

42

51

Fulfilled

Hardly ever/
Never
33

58

Frustrated

Angry

Sometimes

49

29

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

31
PEW RESEARCH CENTER

pride in their work, only 37% also say they regularly feel frustrated by their job.
Roughly four-in-ten officers say their work makes them feel fulfilled nearly always (9%) or often
(33%), while about half as many say their work makes them angry (3% nearly always, 19% often).
Feelings about police work vary significantly across key demographic groups. White officers are
significantly more likely than black officers to say they nearly always or often feel frustrated by
their work. Some 54% of white officers say this, compared with roughly four-in-ten (41%) black
officers. Hispanic officers fall in the middle on this measure.
About one-in-four white officers (23%) say their work in law enforcement nearly always or often
makes them feel angry. A smaller share of black officers – 17% – say this. Among Hispanic officers,
21% say they nearly always or often feel angry, a share that is not significantly different from white
or black officers.

Race and rank linked to feelings about police work
% of officers saying their work as a law enforcement officer nearly always or often makes them feel …

Note: Whites and blacks include only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

When it comes to positive feelings about police work, white, black and Hispanic officers are more
closely aligned. Similar shares of each group of officers say their work in law enforcement nearly
always or often makes them feel proud and fulfilled (though the share of Hispanic officers who feel
this way is slightly higher than the share of whites).

www.pewresearch.org

32
PEW RESEARCH CENTER

Police officers who are new to the job are more likely than more seasoned officers to say their work
makes them feel proud or fulfilled, and they are less likely to say they often feel frustrated. Sevenin-ten officers who have been on the job less than five years say they nearly always or often feel
proud of the work they do. By comparison, 57% of those who have been on the job five years or
more say they always or often feel proud. Similarly, 59% of new officers say their work makes them
feel fulfilled nearly always or often, a sentiment shared by 40% of officers who’ve been on the force
for five years or more. And while 39% of new officers say they nearly always or often feel frustrated
by their job, 53% of those with five or more years on the force say this.
There are sharp gaps by rank in the emotions officers experience on the job. Administrators (73%)
are much more likely than rank-and-file officers (57%) or sergeants (59%) to say they nearly
always or often feel proud of the work they do. Administrators also say they more often feel
fulfilled by their job: 54% of administrators vs. 41% of rank-and-file officers and 40% of sergeants
say their work nearly always or often makes them feel fulfilled.
In addition, administrators are less likely to voice frustration or even anger over their jobs. While
43% of administrators say their job nearly always or often makes them feel frustrated, about half
of rank-and-file officers (52%) and sergeants
(53%) say they often feel this way. Rank-andWhite officers differ from blacks and
file officers (23%) and sergeants (24%) are
Hispanics over when it’s appropriate to
roughly twice as likely as administrators (13%)
break department rules
to say their work nearly always or often makes
% of officers saying they would advise another officer to
them feel angry.

For police, sometimes moral
imperative trumps department
rules
Sometimes police are faced with situations
where doing what is morally the right thing
would require breaking a department rule. By
a ratio of 57% to 40%, officers say they would
advise a fellow officer in this type of situation
to do the right thing rather than follow the
rule.
The balance of opinion on this question is
fairly consistent across major demographic

___ if faced with a situation in which doing the morally
right thing required breaking a department rule
Do the morally
right thing
All officers

Whites
Blacks
Hispanics

57

Follow
department
rule
40

63

34
43
47

55
51

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

33
PEW RESEARCH CENTER

groups, with one exception. White officers have a much different view on this than their black and
Hispanic counterparts.
Among white officers, 63% say they would advise another officer to do the morally right thing,
even if it requires breaking a department rule. Only 43% of black officers and 47% of Hispanic
officers say the same. Among black officers, 55% say they would advise a fellow officer to follow
the department rule.
Views on this hypothetical dilemma vary only modestly across rank and job tenure. Rank-and-file
officers, sergeants and administrators all lean narrowly toward advising a fellow officer to do the
morally right thing even if it requires breaking a department rule. Administrators are slightly more
likely than the rank and file to say this.
Officers who work in urban departments (60%) are somewhat more likely than those working in
suburban departments (53%) to say they would advise a colleague to do the morally right thing
even if it meant breaking the rules.
Police were also asked about the extent to which the so-called code of silence prevails in their
department when someone witnesses wrongdoing or unethical behavior by a fellow officer.
Respondents were asked to imagine a scenario in which an officer who is on duty is driving his
patrol car on a deserted road and sees a vehicle that is stuck in a ditch. The officer discovers that
the driver is a fellow officer who is not hurt but is obviously intoxicated. Instead of reporting the
accident and the offense, he drives the intoxicated officer home. (The scenario did not include any
information about the potential punishment for the intoxicated officer.)
After being presented with this set of facts, roughly half (53%) of the officers surveyed said that
most officers in their department would not report the officer who covered up for his colleague,
including about a quarter who say only a few (22%) or none (5%) of their peers would report the
cover-up.
On the flip side, roughly three-in-ten officers (29%) surmised that all or most of the officers in
their department would turn in the officer who covered up. Some 15% said about half of the
officers in their department would turn in their colleague.

www.pewresearch.org

34
PEW RESEARCH CENTER

2. Inside America’s police departments
Most police officers are satisfied with their department as a place to work and are strongly
committed to ensuring their agency is successful. But police offer less positive views about some
key aspects of their department’s processes and policies. For example, officers are divided on
whether the disciplinary process in their department is fair. And a majority of officers do not feel
that officers who are consistently doing a poor job are held accountable.
Police also indicate that their department does not have sufficient resources or training. A vast
majority of officers say their department does not have enough officers to adequately police the
community. And only about three-in-ten (31%) officers say their department has done very well in
equipping them adequately to do their job. Roughly four-in-ten officers say their department has
done very well in training them adequately to perform their job (39%) and communicating their
job responsibilities clearly (37%).
The survey also asked officers about their views on the use-of-force guidelines in their department.
About a quarter of officers (26%) say that their department’s use-of-force guidelines are too
restrictive – 73% say the guidelines are about
right. But only about a third of officers (34%)
A majority of officers are satisfied with
say that their department’s use-of-force
their agency as a place to work
guidelines are very useful when they are
% of officers saying they are ___ with their agency as a
confronted with situations where force may be
place to work
necessary.
Very
satisfied

When it comes to preventing the use of
unnecessary force, an overwhelming majority
say officers should be required to intervene
when they think another officer is about to use
unnecessary force.
This chapter examines how police officers
across the nation view the leadership and the
internal processes and policies within their
own departments. It also explores officers’
concerns about how their colleagues interact
with the public.

Satisfied

NET

All officers

16

58

74

Rank-and-file
Rank-andofficers
file
officers

15

58

73

Sergeants

15

57

72

Administrators

36

52

88

Note: NETs calculated before rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

35
PEW RESEARCH CENTER

Most officers are satisfied with their department and committed to its
success
Roughly three-quarters (74%) of police officers say they are satisfied with their agency as a place to
work, although relatively few (16%) say they are very satisfied. Levels of satisfaction with their
department vary by officer’s rank. While majorities of each rank say they are at least satisfied with
their agency as a place to work, administrators are about twice as likely to say that they are very
satisfied with their agency (36% vs. 15% of rank-and-file officers and sergeants).
Newer officers (25%) are also more likely to express a high level of satisfaction with their
department than those who have been in the field for five or more years (15%).
Overall, officers express a firm commitment to their department: Almost all officers strongly agree
(60%) or agree (36%) that they are strongly committed to making their agency successful. This is
consistent across all demographic groups and agency characteristics.

Support for top leadership’s direction is stronger in smaller agencies
Police offer mixed support for their top
leadership. Three-in-ten officers say they are
extremely or very supportive of the direction
their top management is taking the
organization, and 28% say they are moderately
supportive. Roughly a third (34%) of officers
say they are slightly supportive or not
supportive at all.

Police offer mixed support for their
agency’s top leadership
% of officers saying they are ___ supportive of the
direction that top management is taking their
organization
Extremely/Very
All officers

Police in agencies with fewer than 1,000
officers are more supportive of their
leadership’s direction than those in larger
agencies. About four-in-ten (39%) officers at
agencies with fewer than 1,000 officers say
they are extremely or very supportive of their
management’s direction. Only 20% of officers
in departments with at least 1,000 say the
same; 44% of officers in these departments say
they are slightly or not at all supportive of
their top management’s direction.

Moderately

30

Slightly/Not at all

28

34

Agency size
<1,000 officers
1,000 or more

39
20

29
28

25
44

Note: “Don’t know management’s direction” and no answer
categories not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

36
PEW RESEARCH CENTER

Most officers feel their supervisor treats them and their colleagues with respect
About four-in-ten (43%) officers say their
supervisor always respects them and their
colleagues, while about as many (41%) say this
is usually the case; 15% say their supervisor
sometimes, hardly ever or never treats the
officers he or she supervises with respect.
When asked how often employees in their
department are asked for input on decisions
that will affect them, roughly half (54%) of
officers say this is hardly ever (39%) or never
(15%) the case. Still, 45% of officers say they
are at least sometimes asked for input about
decisions that will affect them.
Women are more likely than men to say
employees are hardly ever or never asked for
their input. Among female officers, 63% say
that they and their co-workers are hardly ever
(45%) or never (18%) asked for their input. By
comparison, 53% of men say the same
(comprising 38% who say hardly ever and 15%
who say never).
Police in large departments (with 2,600
officers or more) are about three times as
likely as those in small departments to say
they are never asked for their input. Just 8% of
police in departments of fewer than 500
officers say they are never asked for input on
decisions that will affect them. By comparison,
27% of police in departments with at least
2,600 officers say the same.

Most officers say they are treated with
respect by their supervisor
% of officers saying their supervisor ___ treats officers
he or she supervises with respect
Always

Usually

Sometimes

43

Hardly ever

Never
11 4 1

41

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

About half of officers say they are hardly
ever or never asked for input about
decisions that will affect them
% of officers saying that employees in their departments
are ___ asked for their input on decisions that will affect
them
Always/Usually

Sometimes

Hardly ever

Never

All officers 11

34

39

15

Men

35

38

15

12

Women 8

18

45

30

Agency size
<500 officers

15

500 to 2,599

13

2,600 or more 6

11

40

36
22

8

33

43

44

27

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

37
PEW RESEARCH CENTER

For promotions and assignments, minorities and women are more likely
to say their counterparts are treated better
In America’s police departments, women and Hispanics are underrepresented despite their
growing shares in recent decades. In 2013 (the most recent data available), blacks made up the
same share of local police officers as they did in the U.S. adult population. (See “Growing diversity
inside America’s police departments” textbox for more details on diversity in U.S. police
departments.) The survey asked how women and minority officers are treated relative to their
counterparts when it comes to assignments and promotions in their department.
Some 56% of officers say that minorities and whites are treated about the same way when it comes
to assignments and promotions, while about three-in-ten (31%) say minorities are treated better
than whites in these cases. Roughly one-in-ten (11%) officers say that whites are treated better
than minorities when it comes to assignments
and promotions in their department.
About half of black officers say whites
But officers of different racial and ethnic
backgrounds offer vastly different views on
how minorities are treated in comparison to
whites when it comes to assignments and
promotions. Roughly six-in-ten white (61%)
and Hispanic (58%) officers say that
minorities and whites are treated about the
same. By contrast, about half (53%) of black
officers say that whites are treated better than
minorities, while about one-in-five (19%)
Hispanic officers and only 1% of white officers
say the same.
About four-in-ten (37%) white officers say that
minorities are treated better than white
officers when it comes to decisions about
assignments and promotions. By comparison,
23% of Hispanic officers and 6% of black
officers say minorities are treated better than
whites. 7

are treated better than minorities in
assignments and promotions

Q. When it comes to decisions about assignments and
promotions, which comes closest to describing how
things work in your department? (%)
Whites are
treated better
than minorities
All officers

Minorities and
whites are treated
about the same

37

61

Blacks
Hispanics

31

56

11

Whites 1

Minorities are
treated better
than whites

19

6

39

53
58

23

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

7

www.pewresearch.org

38
PEW RESEARCH CENTER

Growing diversity inside America’s police departments
America’s police departments have become increasingly diverse since the late 1980s. According to the Bureau
of Justice Statistics, about 504,000 sworn police officers were employed in local U.S. police departments in
2013, the most recent year for which data were available.7 Among full-time officers, 12% were women in 2013,
up from 8% in 1987. Women accounted for about one-in-ten supervisory or managerial positions and 3% of
local police chiefs in 2013 (these data were collected for the first time in 2013). In general, departments that
serve larger populations tend to have a higher share of women on their police force and a higher share of
female supervisors.
Police departments have also become more racially and ethnically diverse: In 2013 more than a quarter (27%)
of officers were racial or ethnic minorities, up from 23% in 2000. Some 12% of full-time local police officers in
2013 were black, equal to their share of all U.S. adults. Another 12% of full-time officers were Hispanic,
compared with 15% of U.S. adults in 2013. While the share of police officers who are black has remained
largely the same since 2000, the share who are Hispanics has grown by about 3 percentage points.8

Most sworn officers in local police departments were full-time employees (477,317). This figure does not include officers in sheriff’s
departments or state agencies.
8 Change calculated prior to rounding.
7

www.pewresearch.org

39
PEW RESEARCH CENTER

Roughly four-in-ten female officers say men are treated better when it comes to
assignments and promotions
More than half (59%) of officers say that when
it comes to assignments and promotions, men
and women are treated the same in their
department, while about three-in-ten (28%)
say women are treated better than men. Some
12% say that men are treated better than
women in these cases.
Views on how men and women are treated in
their department vary greatly by gender.
About four-in-ten (43%) women say that men
are treated better than women when it comes
to assignments and promotions, but only 6%
of men say the same. And a third of men say
that women are treated better in these cases,
compared with just 6% of women. Six-in-ten
men and half of women say they are treated
about the same when it comes to assignments
and promotions.

For assignments and promotions, about
four-in-ten female officers say men are
treated better than women
Q. When it comes to decisions about assignments and
promotions, which comes closest to describing how
things work in your department? (%)
Women are
treated better
than men

Men and women
are treated
about the same

Men are
treated better
than women
Men 6

33

60

Women

6

50

43

Among men
Whites 3
Blacks

32

64

24

53

21

Hispanics 6

39

54

Among women

Black men and women are more likely than
their white counterparts to say that men are
treated better than women in their
departments. Among men, 21% of black
officers and 3% of white officers say this. Just
6% of Hispanic men say that men are treated
better than women.

Whites
Blacks
Hispanics

34

61
44

5

62

33

48

5
6

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”

Among women, black officers are about twice
PEW RESEARCH CENTER
as likely as white officers to say that men are
treated better than women when it comes to
assignments and promotions (61% vs. 33%). Some 44% of Hispanic women say this is the case.

It’s worth noting, however, that across racial and ethnic groups, female officers are far more likely
than their male counterparts to say men in their departments are treated better than women when

www.pewresearch.org

40
PEW RESEARCH CENTER

it comes to assignments and
promotions. Men across all
major racial and ethnic
groups are more likely than
their female counterparts to
say that women are treated
better than men.

Officers are divided
on the fairness of
their agency’s
disciplinary process

Officers’ assessments of key aspects of the
disciplinary process are mixed
% of officers saying they ___ with each of the following statements
Disagree Agree
The disciplinary process in their
agency is fair
Officers who consistently do a
poor job are held accountable

53

45

72

For minor mistakes, their agency
helps officers with coaching and
counseling rather than punishment

27

40

59

Police offer mixed
Note: “Agree” comprises those who say they agree or strongly agree with the statement.
“Disagree” comprises those who say they disagree or strongly disagree with the statement.
assessments on some key
No answer category not shown.
aspects of the disciplinary
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
process in their departments.
PEW RESEARCH CENTER
When asked if they agree or
disagree with the statement
that the disciplinary process in their agency is
For minor mistakes, more officers today
fair, officers are divided: 45% say they strongly
agree coaching is used vs. punishment
agree or agree with this statement; 53% say
% of officers saying they ___ that, for minor mistakes,
they disagree or strongly disagree.
their department helps officers with coaching and
counseling rather than punishment

But a majority (72%) of officers say they
disagree (47%) or strongly disagree (25%) that
officers in their department who consistently
do a poor job are held accountable. Roughly a
quarter (27%) of officers agree, including just
3% who strongly agree. While similarly large
shares of officers across all demographic
groups disagree that officers who consistently
do a poor job are held accountable, police in
larger agencies with at least 1,000 officers are
particularly likely to disagree (81% vs. 63% of
police in departments with fewer than 1,000
officers).

55

44

Agree
52
47
Disagree

July Nov. 2013

Oct. 2014 Feb. 2015

59

40

May Aug. 2016

Note: “Agree” comprises those who say they agree or strongly agree
with the statement. “Disagree” comprises those who say they
disagree or strongly disagree with the statement. No answer
category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

41
PEW RESEARCH CENTER

Police offer a slightly more positive view when asked whether they agree or disagree that their
department helps officers with coaching and counseling rather than punishment for minor
mistakes: About six-in-ten (59%) agree, while 40% disagree. Officers in departments of fewer than
1,000 officers are more likely than those in larger departments to agree that their department
helps officers with coaching and counseling rather than punishment for minor mistakes (69% vs.
49%). Officers of higher rank are also more likely to say this is the case (79% of administrators
agree vs. 57% of rank-and-file officers).
The share of officers who agree that for minor mistakes the department helps officers with
coaching and counseling rather than
punishing them is up 7 percentage points since
Officers in smaller agencies more likely
it was last asked about two years earlier
to agree that their agency’s disciplinary
(October 2014 to February 2015). At that time,
process is fair
officers were divided on the topic: 52% said
% of officers saying they ___ that the disciplinary
they agreed with the statement that in their
process in their agency is fair
department officers are helped with coaching
and counseling rather than punished for
Agree
Disagree
minor mistakes, while 47% disagreed.
All officers
45
53

Men and white officers more likely than
counterparts to agree that their agency’s
disciplinary process is fair
Views of the disciplinary process being fair
vary by the characteristics of the officers and
the department. Men are more likely than
women to say they agree that the disciplinary
process in their agency is fair (47% vs. 39%).

Men
Women

Hispanics

39

59

Whites
Blacks

47

52

50

49

41

58

36

62

Agency size

While half of whites agree that the disciplinary
process in their agency is fair, 41% of black
officers and 36% of Hispanic officers say the
same.
Police in smaller agencies are about twice as
likely as those in larger agencies to agree that
the disciplinary process in their department is
fair. About six-in-ten (59%) of those in

<1,000 officers
1,000 or more

40
67

59
31

Note: “Agree” comprises those who say they agree or strongly agree
with the statement. “Disagree” comprises those who say they
disagree or strongly disagree with the statement. No answer
category not shown. Whites and blacks include only non-Hispanics.
Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

42
PEW RESEARCH CENTER

agencies with fewer than 1,000 officers agree with the statement, while about three-in-ten (31%) in
agencies with 1,000 or more officers agree.

A majority of officers say there are
not enough police in the
community where they work
Fully 86% of officers say their department
does not have enough officers to adequately
police the community. Similarly large shares of
officers say this across all demographic groups
and agency characteristics. But police in
agencies with fewer than 1,000 officers are
about four times as likely as those in larger
agencies to say they do have enough officers to
police the community (21% vs. 5%).

Most officers say their department has
too few officers to police the community
% of officers saying their department ___ have enough
officers to adequately police the community
Does not
All officers

Does

86

13

Agency size
<1,000 officers

79

21

1,000 or more 95

5

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”

Officers of higher rank are also more likely to
say that their department has enough officers.
PEW RESEARCH CENTER
About a quarter (23%) of administrators say
their department has enough officers to
adequately police the
community, compared with
About four-in-ten officers say their department has
13% of rank-and-file officers
done very well in training them adequately for their job
and 10% of sergeants.
% of officers saying their department has done each of the following …
The survey also asked officers
how well their department
has trained and equipped
them. Roughly four-in-ten
officers say that their
department has done very
well in training them
adequately to do their job
(39%) and in communicating
their job responsibilities to
them clearly (37%). And
about three-in-ten (31%) say

Very
well

Somewhat
well

Not too
well

Not at all
well

Train them adequately for
their job

39

42

14

5

Communicate their job
responsibilities to them
clearly

37

44

13

5

Equip them adequately to
perform their job

31

44

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

17

8

43
PEW RESEARCH CENTER

the department has done very
well in equipping officers to
adequately perform their job.
In all, a majority of officers
say that their department has
done at least somewhat well
in each of these areas.

Officers in larger agencies less likely to say their
department has trained and equipped them very well
% of officers in departments with ____ officers saying their department has
done each of the following very well
<1,000
Training them adequately
for their job

1,000 or more
49
29

Police in larger agencies are
45
Communicating their job
considerably less likely to say
repsonsibilities clearly
29
their department has done
41
very well in each of these
Equipping them adequately to
perform
theirjob
19
aspects. For example, 29% of
police in agencies with at
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
least 1,000 officers say their
“Behind the Badge”
department has done very
PEW RESEARCH CENTER
well in training them
adequately for their job,
compared with 49% of police in agencies with fewer than 1,000 officers. Likewise, just 19% of
police in agencies with 1,000 or more officers say their department has done very well in
equipping them adequately to perform their job, compared with 41% of those in smaller agencies.
Officers with less than five years of experience are more likely than those with more experience to
say their department has done very well in each of these areas. For example, 57% of officers with
less than five years of experience say their department has done very well in training them
adequately for their job, compared with 37% of officers with five or more years of experience.

About a third of officers say their department’s use-of-force guidelines
are very useful
When asked whether their department’s use-of-force guidelines are too restrictive, not restrictive
enough or about right, a majority (73%) of officers say that their department’s rules are about
right. Still, about a quarter (26%) say the use-of-force guidelines are too restrictive. Only about 1%
say their department’s guidelines are not restrictive enough.
While similar shares of officers across demographic groups say their use-of-force rules are about
right, police in larger agencies with at least 1,000 officers are about twice as likely as officers in

www.pewresearch.org

44
PEW RESEARCH CENTER

smaller agencies to say that their department’s
rules governing the use of force are too
restrictive (37% vs. 15%).
Rank-and-file officers and sergeants are
somewhat more likely than administrators to
say their department’s rules governing the use
of force are too restrictive. Roughly a quarter
of rank-and-file officers (27%) and sergeants
(24%) say this is the case, compared with 17%
of administrators.
Roughly a third (34%) of officers say that their
department’s use-of-force guidelines are very
useful when they are confronted with actual
situations where force may be needed. An
additional 51% say they are somewhat useful.
Some 14% say they are not too useful or not at
all useful.
Those in agencies with fewer than 1,000
officers are considerably more likely than
those in larger agencies to say their agency’s
guidelines are very useful. About four-in-ten
(41%) in agencies with fewer than 1,000
officers say this, compared with 27% in
agencies with at least 1,000 officers.
Administrators are also substantially more
likely than rank-and-file officers to say their
department’s guidelines are very useful (49%
vs. 33%, respectively). 8

Officers in larger departments are more
likely to say their use-of-force guidelines
are too restrictive
% of officers saying that in general the rules governing
the use of force in their departments are …
Too restrictive
All officers

About right

26

73

Agency size
<1,000 officers

15

1,000 or more

84
37

62

Note: “Not restrictive enough,” “The department has no such
guidelines” and no answer categories not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

About a third of officers say, when put to
the test, use-of-force guidelines in their
department are very useful
% of officers saying their department’s use-of-force
guidelines are ___ when officers are confronted with
actual situations where force may be needed
Very useful
All officers

34

Somewhat useful
51

NET
85

Agency size
<1,000 officers
1,000 or more

41
27

50
52

91
79

Note: NETs calculated before rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
8

PEW RESEARCH CENTER

www.pewresearch.org

45
PEW RESEARCH CENTER

Police worry more that their fellow officers will spend too much time versus not enough
time diagnosing a situation before acting
More police officers worry about their fellow
officers spending too much time diagnosing a
situation before acting (56%) than about their
fellow officers not spending enough time
before acting decisively (41%).
Black officers and administrators stand out as
the only groups studied that are more likely to
say they worry more that officers in their
department will not spend enough time
diagnosing the situation before acting than
that they will spend too much time.
About six-in-ten (61%) black officers say they
worry more that officers will not spend enough
time diagnosing the situation before acting,
compared with 37% of white officers and 44%
of Hispanic officers. And 59% of
administrators say the same. By comparison,
just 40% of rank-and-file officers and
sergeants say they worry more that officers
will not spend enough time diagnosing the
situation before acting.

More police worry their fellow officers
will not act quickly enough than worry
they will act too quickly
% of officers saying that when it comes to the way most
officers in their department deal with members of the
public, they worry more that the officer will ___
diagnosing the situation before acting decisively
Not spend
enough time

Spend too
much time

All officers

41

56

Men

40

57

Women
Whites
Blacks

48

49

60

37
37

61

Hispanics

53

44

Rank-and-file officers

40

57

Sergeants

40

58

Administrators
Years in law enforcement
Less than 10
10 to 19
20 or more

39

59
37
39
47

60
58
50

Men are more likely than women to say that
Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
officers in their department will spend too
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
much time diagnosing the situation before
“Behind the Badge”
acting (57% vs. 48%, respectively). And about
PEW RESEARCH CENTER
six-in-ten officers with less than 20 years of
experience say they worry more that officers
will spend too much time diagnosing a situation before acting, compared with half of officers with
20 or more years of experience.

www.pewresearch.org

46
PEW RESEARCH CENTER

Most officers say that they should be required to intervene when another officer is about to
use unnecessary force
One policy recommendation to protect
community members and police from
unnecessary force is to require police to
intervene when they think another officer may
use unnecessary or excessive force. 9 A majority
(84%) of police say that officers should be
required to intervene when they believe
another officer is about to use unnecessary
force, while just 15% say they should not be
required to intervene.
Majorities of police officers across all
characteristics say that officers should be
required to intervene in these scenarios. But
rank-and-file officers are more likely than
those of higher ranks to say they should not be
required to intervene when they think another
officer is about to use unnecessary force (17%
of rank-and-file officers say this, compared
with 9% of sergeants and 5% of
administrators).

Most officers favor a requirement to
intervene when another officer is about
to use unnecessary force
% of officers saying that officers ____ to intervene when
they believe another officer is about to use unnecessary
force
Should not
be required

Should be
required

All officers 15

84

Rank-and-file officers 17

82

Sergeants
Administrators

9
5

91
95

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

About four-in-ten officers say they are expected to meet a quota for
arrests or tickets
In recent years quotas – which have been used to measure performance for officers – have been
criticized for a range of reasons related to their limits in assessing the quality of policing. And
today, in several states ticket and arrest quotas are illegal.
While few officers (3%) say that they are formally expected to meet a predetermined number of
tickets, arrests, citations or summonses in their unit, about a third (34%) of officers say there are
informal expectations for meeting a predetermined number of arrests or tickets. A majority (63%)
say officers in their unit are not expected to meet any predetermined number of tickets or arrests.
In 2016, the Police Executive Research Forum, a nonprofit focused on critical issues in policing, developed 30 guiding principles on use of
force to help protect officers and the communities they serve. The requirement for intervening to prevent other officers from using excessive
force was one of the recommended policies.
9

www.pewresearch.org

47
PEW RESEARCH CENTER

Rank-and-file officers – those who routinely
monitor an area, issue tickets and make
arrests – are particularly likely to say there are
informal expectations. Some 36% of rank-andfile officers say there is an informal
expectation for meeting a predetermined
number of tickets or arrests; 29% of sergeants
and 23% of administrators say the same.

About four-in-ten officers say they are
formally or informally expected to meet
a certain number of arrests or tickets
% of officers saying that officers in their unit have ___
to meet a predetermined number of tickets, arrests,
citations or summonses
A FORMAL
expectation
3

34

An INFORMAL
expectation

No expectation

63

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

48
PEW RESEARCH CENTER

3. Police and the community
Police officers routinely deal
with people behaving at their
worst. Frequent encounters
with verbally abusive and
sometimes physically
combative citizens also come
with the badge.

Most officers say the public respects them, few feel
distrustful of most people
% of officers saying they __ with each of the following statements
Strongly disagree

Disagree

Agree

Strongly agree

NET 68%
Most people
respect the
police

7
25
61
6
Despite these experiences, the
Pew Research Center survey
NET 28%
finds that a majority of
Officers have
officers retain a generally
reason to be
16
56
23
5
distrustful of
positive view of the public.
most citizens
About seven-in-ten reject the
Note: No answer category not shown. NET calculated before rounding.
assertion that most people
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
can’t be trusted, and a similar
PEW RESEARCH CENTER
share believes that most
people respect the police.
These opinions, if anything, have grown somewhat more positive in recent years despite the
national outcry over police methods and behaviors that followed a series of recent, highly
publicized deaths of black men at the hands of law enforcement officers.

Rather than viewing the neighborhoods where they work as hostile territory, about seven-in-ten
officers say at least some or most of the residents share their values. More than nine-in-ten believe
it is important for an officer to know the people, places and the culture in the areas where they
work in order to be effective at their job.
About nine-in-ten officers (91%) also say police have an excellent or good relationship with whites
in their communities. But just 56% rate the relationship between police and blacks positively,
while seven-in-ten report good relations with Hispanics. These perceptions differ dramatically
depending on the race or ethnicity of the officer. For example, six-in-ten white officers characterize
police relations with blacks in their areas as excellent or good, a view shared by only 32% of black
officers.
The survey also finds that officers are divided over the use of more aggressive and potentially more
controversial methods to deal with some people or to use in some neighborhoods in their

www.pewresearch.org

49
PEW RESEARCH CENTER

communities. A modest majority (56%) agree that aggressive tactics are more effective than a
more courteous approach in certain areas of the city, but 44% disagree with this premise. Another
44% agree or strongly agree that some people can only be brought to reason the hard, physical
way. Younger, less experienced and lower ranking officers are significantly more likely to favor
these more confrontational approaches than older, more experienced department administrators.
The survey also finds that police work takes an emotional toll on many officers. A 56% majority say
they have become more callous toward people since they started their job. This perceived change
in outlook is closely linked to increased support for aggressive or physically punishing tactics. In
addition, officers who say they have become more callous on the job report significantly higher
levels of work-related anger and frustration than other officers. They also are more likely to have
fought or struggled with a suspect who
was resisting arrest in the past month or
Rank-and-file officers less likely than
to have fired their service weapon
police supervisors or administrators to
sometime in their career.
have a positive view of the public

Police feel respected

% of officers saying they __ that most people respect the
police

Roughly two-thirds of all officers agree
(61%) or strongly agree (6%) that most
people respect the police. About seven-inten (72%) reject the statement that
“Officers have reason to be distrustful of
most citizens.”
Comparisons with earlier surveys by the
National Police Research Platform
(NPRP) find that these views of the public
have not grown more negative in the wake
of recent deadly encounters involving
police and black men. If anything, these
data suggest police views of the public
have gotten more favorable in the past
year and a half.
In the NPRP survey conducted in
September 2013 to January 2014, six-inten officers said most of the public

Agree

Disagree
Rank-and-file
officers
Sergeants

35

65

29

71

Administrators 14

86

% of officers saying they __ that officers have reason
to be distrustful of most citizens

Rank-and-file
officers
Sergeants
Administrators

Disagree

Agree

70

30

75
86

25
14

Note: No answer category not shown. “Agree” comprises those who
say they agree or strongly agree with the statement. “Disagree”
comprises those who say they disagree or strongly disagree with the
statement.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

50
PEW RESEARCH CENTER

respects the police. A somewhat smaller share (55%) expressed the same opinion in a NPRP survey
conducted in October 2014 to February 2015, months after the Michael Brown shooting. But since
then, the share who says police are respected has rebounded to 68%.
The mistrust measure has varied less in recent years. In the 2013-14 survey, 67% of officers
disagreed that officers have reason to be distrustful of most citizens, a view shared by 69% of
officers in 2014-15 and 72% in the latest survey.
These views differ significantly by rank. Rank-and-file officers – a group largely composed of the
men and women with the greatest contact with average citizens – have a significantly less
favorable view of the public than do administrators. About two-thirds of rank-and-file officers
(65%) but 86% of administrators believe that most people respect the police.
Similarly, 70% of rank-and-file officers but
86% of administrators disagree or strongly
disagree that police have reason to distrust
most people.

Police see need to understand the
community
Most officers agree that in order to be
effective, police need to understand the people
in the neighborhoods they patrol. About
seven-in-ten (72%) say it is very important for
an officer to have detailed knowledge of the
people, places and culture in the areas where
they work, while a quarter say it is somewhat
important. Only 3% say knowledge of the
neighborhoods they patrol is not too or not at
all important.
Yet the degree to which officers value local
knowledge varies significantly by the officer’s
race and gender. Fully 84% of black officers
and 78% of Hispanics say knowledge of the
people, places and culture of the
neighborhoods they patrol is very important to

Most officers say police need to know
the community to be effective
% of officers saying it is __ for officers to have detailed
knowledge of the people, places and culture in the areas
where they work
Very important
Somewhat important
Not too important/Not at all important
All officers

72

25

3

Men

71

26

3

Women

Whites
Blacks
Hispanics

80

18

3

28

69
84
78

2

14
20

2
2

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

51
PEW RESEARCH CENTER

be effective at their job, a view shared by 69% of whites. Female officers also are more likely than
males to place a premium on local knowledge (80% vs. 71%).

Police say they share values with at least some residents where they work
Overall, about seven-in-ten officers say at least some (59%) or most or nearly all (11%) of the
people in the neighborhoods where they routinely work share their values and beliefs.
Significant differences emerge when these results are broken down by the officer’s rank: About
two-thirds of rank-and-file officers (68%) believe that some or most of the people living in their
patrol areas share their beliefs. By contrast, three-quarters of sergeants and 85% of administrators
say the same thing.
When the analysis is limited to rank-and-file officers – the group that arguably has the most direct
daily contact with citizens – views differ significantly within key demographic groups. Most
notably, younger rank-and-file officers and
those in larger departments are less likely than
Younger rank-and-file officers less likely
older officers or those in small police
to say the people in the neighborhoods
departments to say they share common values
where they work share their values
and beliefs with at least some of the people in
% of rank-and-file officers saying __of the people in
the areas they patrol.
About six-in-ten rank-and-file officers (62%)
ages 18 to 34 say some or most of the people in
the neighborhoods where they work share
their beliefs and attitudes. By contrast, about
three-quarters (76%) of rank-and-file officers
ages 50 and older express a similar view.
Rank-and-file officers in larger departments
also are less likely to share values with the
people in the areas where they patrol. Eightin-ten rank-and-file officers working in
departments with fewer than 300 sworn
personnel say they share values and beliefs
with at least some of the people they patrol. By
contrast, about six-in-ten (62%) rank-and-file

the neighborhoods they routinely patrol share their
values and beliefs
Very few/None
Rank-and-file
officers

31

All or most/Some
68

Among rank-and-file officers ...
Ages 18-34

38

62

35-44

30

45-49

29

50+

24

69
71
76

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

52
PEW RESEARCH CENTER

officers in departments with 2,600 or more
sworn personnel say the same. (This difference
may not be surprising. Larger departments
typically serve urban areas with a more diverse
set of neighborhoods than smaller
communities. These urban neighborhoods
often can be home to various nationalities and
racial, ethnic, language and religious groups
with attitudes and beliefs that may be very
different from those of the rank-and-file
officer. 10)
About half or more of all officers say their
departments have excellent or good relations
with major racial and ethnic groups in the
communities where they work. This overall
positive assessment varies considerably by the
racial/ethnic group and also by the race and
ethnicity of the officer. Black officers in
particular are significantly less likely than
white or Hispanic officers to rate relations
with minority groups in their community
favorably. (Note: The percentages are based on
only those officers who offered a rating.)
Overall, about nine-in-ten officers (91%)
characterize relations between police and
whites in their communities as excellent (22%)
or good (69%). By contrast, 56% of all officers
have a similarly positive view of relations
between police and the black community (8%
say relations are excellent, while 47% say they
are good). Seven-in-ten say relations with
Hispanics are positive, and 88% say the same
about Asians.

About half or more officers say police
have positive relations with the racial,
ethnic groups in their communities
% of officers saying they would rate relations between
the police in their department and the following groups
in the community they serve …
Excellent

Good

Only fair

Poor

Whites in the community
NET 91%
22

69

8

1

Blacks in the community
NET 56%
8

47

26

18

Hispanics in the community
NET 70%
10

60

22

7

Asians in the community
NET 88%
18

70

11

2

Note: NET calculated before rounding. Percentages based on those
officers who rated relations with each group. A total of 15% of all
officers report they could not rate relations with Asians because
there are too few in the community where they worked; 2% say
there are too few blacks and 3% say there are too few Hispanics or
whites.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

For an analysis of the changing demographics of metropolitan, suburban and exurban areas, see Frey, William H. May 2011. “Melting Pot
Cities and Suburbs: Racial and Ethnic Change in Metro America in the 2000s.” Brookings Institution.
10

www.pewresearch.org

53
PEW RESEARCH CENTER

Black officers see police-minority relations less positively
About nine-in-ten white, black and Hispanic officers agree that police and whites in their
communities have good relations. But striking differences emerge when the focus shifts to how
black, white and Hispanic officers view police-minority relations in their communities.
Only about a third of all black officers (32%) say relations between police and blacks in their
community are excellent or good, while about twice as many (68%) characterize police-black
relations as only fair or poor.
By contrast, six-in-ten white and Hispanic
officers report that police-black relations in the
communities they serve are excellent or good.
Views also diverge along racial lines when the
focus turns to how black, white and Hispanic
officers view police-Hispanic relations. Roughly
three-quarters of white officers (76%) and 71% of
Hispanic officers say police in their communities
have excellent or good relations with Hispanics.
By contrast, only 46% of black officers share that
positive assessment, while 54% characterize
relations between police and Hispanics as only
fair or poor.
A similar but more muted pattern is apparent on
views of police relations with Asians in their
community. About nine-in-ten white and
Hispanic officers (91% and 88%, respectively)
say relations between police and Asians are
excellent or good, while 75% of black officers
agree.

Use of aggressive, physical tactics
To measure the extent to which officers endorse
the use of aggressive tactics in some situations
over less potentially provocative techniques, the
survey asked officers how much they agreed or

www.pewresearch.org

54
PEW RESEARCH CENTER

disagreed with two
statements. The first
statement read, “In certain
areas of the city it is more
useful for an officer to be
aggressive than to be
courteous.” The second
measured support for the
assertion that “some people
can only be brought to reason
the hard, physical way.”

Some officers say tough, aggressive tactics are
needed with some people and in some neighborhoods
% of officers saying they __ with each of the following statements
Strongly disagree

Disagree

Agree

Some people can only be
brought to reason the hard, 10
physical way
In certain areas of the city it
is more useful for an officer
to be aggressive than to be 10
courteous

Strongly agree
NET 44%

45

5

39
NET 56%

34

39

17

Overall, the survey finds that a
Note: No answer category not shown.
narrow majority (56%) of
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
officers feel that in some
“Behind the Badge”
PEW RESEARCH CENTER
neighborhoods being
aggressive is more effective
than being courteous. A
smaller but still substantial share (44%) agrees or strongly agrees that hard, physical tactics are
necessary to deal with some people, while 55% disagree.
The survey also finds that younger and less senior officers are more likely than older officers or
administrators to favor more potentially provocative methods. About two-thirds (68%) of officers
younger than 35 favor being aggressive over being courteous in some neighborhoods. By contrast,
the share supporting aggressiveness over courtesy falls steadily in each age group to 44% among
officers 50 and older. And while a narrow majority of younger officers (55%) approve of using a
hard, physical approach with some people, support for rough tactics declines to about a third
(36%) for officers 50 and older.
Significant differences in views on both questions emerge when the analytic focus shifts to the
officer’s rank. About six-in-ten rank-and-file officers (59%) support using aggressive tactics in
place of courtesy in some neighborhoods, a view shared by only 34% of department
administrators. To a lesser extent, rank-and-file officers also are more likely than department
administrators to favor harsh, physical methods in dealing with certain people (44% vs. 36%).
Sergeants (46%) also are more likely than administrators to support hard, physical tactics.
The clear differences in the views of lower-ranking officers and more senior administrators raise
this question: Since department administrators are older than rank-and-file officers (median age

www.pewresearch.org

55
PEW RESEARCH CENTER

49 vs. 41), could these differences in attitudes mainly be due to factors associated with officers’
rank or tenure and not their age?
The answer is no. When only
the views of rank-and-file
officers are examined, the
same age pattern is evident:
Fully 69% of rank-and-file
officers younger than 35
favored aggressive tactics
over a courteous approach,
compared with 48% of rankand-file officers 50 and older.
Similarly, slightly more than
half (55%) of rank-and-file
officers under the age of 35
agreed that hard, physical
methods are needed for some
people, compared with 35%
of rank-and-file officers 50
and older.

Younger, lower-ranking officers more likely to favor
aggressive and hard, physical approaches
% of officers saying they agree or strongly agree that …
In certain areas of the city it is more
useful for an officer to be aggressive
than to be courteous.
All officers

68

35-44
45-49
50+

59
50

59
49

44

Ages 18-34

55

35-44
45-49
50+

44

Rank-and-file
officers
officers
Sergeants

All officers

56

Ages 18-34

Some people can only be brought to
reason the hard, physical way.

45
40
36

Rank-and-file
officers

44

Sergeants

46

The relationship between
Administrators
34
Administrators
36
support for harsh tactics and
an officer’s years of police
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
experience follows a similar
PEW RESEARCH CENTER
though more complex pattern
because age is closely
correlated with police experience. Once differences by age are accounted for in the analysis, there
are no significant differences in views based on experience. For example, more than half (55%) of
rank-and-file officers younger than 35 with less than 10 years of experience favor harsher
measures for some people – and so does about the same proportion (54%) of those with 10 or
more years of service.

www.pewresearch.org

56
PEW RESEARCH CENTER

A majority of officers become more
callous
Police work can be emotionally difficult, and it
hardens many officers. According to the
survey, a narrow majority of the police (56%)
say they have become more callous toward
people since taking their job, a view that is
significantly more likely to be held by whites
and younger officers than by blacks or older
department members.
Overall, the survey finds that 13% strongly
agree and an additional 43% agree that they
have become more callous toward people since
taking the job. About a third (34%) disagree,
while 9% strongly disagree.
Younger officers are particularly likely to say
they have become more callous, a view shared
by 62% of officers younger than 35 but only
46% of those 50 or older.
The differences are even greater when the
views of black and white officers are
compared. Only about a third (32%) of black
officers but about twice the share of whites
(62%) report they have become more callous
since taking the job.

Majority of officers say they have
become more callous since taking the
job
% of officers saying they __ that they have become more
callous toward people since they took this job
Strongly disagree

Disagree

Agree

NET 56%
9

34

43

13

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Younger officers, whites likelier to
become more callous since joining
department
% of officers saying they agree or strongly agree that
they have become more callous toward people since they
took this job
All officers

56

Ages 18-34

62

35-44

59

45-49

55

50+

Hispanic officers fall between white and black
officers on this question. About half (51%) of
Hispanic officers say they have grown more
callous, a significantly larger share than
among blacks but significantly smaller than
the proportion of whites.

Strongly agree

46

Whites
Blacks
Hispanics

62
32
51

Note: Whites and blacks include only non-Hispanics. Hispanics are
of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

57
PEW RESEARCH CENTER

Comparisons to the results of previous NPRP police surveys suggest little significant variation in
the share of officers who report becoming more callous. In the 2013-14 survey, the number stood
at 53%, while in the 2014-15 poll, 59% reported growing more callous since taking this job
compared with 56% in the latest poll.

Callousness associated with support
for aggressive, physical tactics
The survey finds that officers who feel they have
grown more callous since starting their job are
also more likely to endorse the use of aggressive
or physically harsh tactics in some situations or
in some parts of the community than officers
who say they have not grown more callous.
Officers who say they have grown more callous
are also more likely than their colleagues who
say they have not to say they are frequently
angered or frustrated by their jobs. They also are
more likely to have been involved in a physical or
verbal confrontation with a citizen in the past
month or to have fired their service weapon
sometime in their careers.
About two-thirds (66%) of those who self-report
having become more callous also agree that it is
more useful in certain neighborhoods for an
officer to be aggressive rather than to be
courteous. By contrast, roughly four-in-ten
(43%) of those who have not become more
callous say this. Similarly, about half of officers
(53%) who say they have become more callous
agree or strongly agree that hard, physical
methods are the only way to deal with some
individuals, a view shared by 32% of those who
say they have not become more callous.

www.pewresearch.org

58
PEW RESEARCH CENTER

It is difficult to determine from these data whether increased callousness is a primary cause or a
consequence of feelings of anger or frustration, or the source of attitudes toward aggressive tactics.
It could be that an increasingly callous outlook breeds anger and aggression in some officers. It
could also be that repeated exposure to confrontations with citizens or frustrations on the job leads
an officer to become more unfeeling.
However, the data suggest that these feelings and behaviors are related. For example, the sense of
having become more callous on the job is associated with how these officers feel about their work,
the survey finds. Those who say they have become more callous are about twice as likely as those
who say they have not to say their job nearly always or often makes them feel angry (30% vs. 12%).
They also are far more likely to nearly always or often feel frustrated by their job (63% compared
with 37% among those who say they have not become more callous since taking their job).
By the same token, those who say they have
grown more callous are significantly less likely
than other officers to say their job nearly
always or often makes them feel fulfilled (32%
vs. 55%) and are less likely to say they often
feel proud (50% vs. 69%) about their work.

Increased sense of callousness
associated with involvement in
confrontations
Among officers who say they have/have not become
more callous, the % who also say they …
More callous

Callousness and experiences
An officer’s sense that he or she has grown
more callous on the job also is associated with
a range of experiences on the streets. While
this analysis does not attempt to determine
whether increased callousness is a primary
cause of these behaviors, these data suggest
they are related.
Among those officers who say they have
become more callous toward people, roughly
four-in-ten (38%) also report they had
physically struggled or fought with a suspect
who was resisting in the past month. By
contrast, about a quarter (26%) of those who
say they have not become more insensitive
were involved in a physical altercation during

NOT more callous

... physically struggled or fought with a suspect who was
resisting arrest in the past month
38
26
... were verbally abused by a community member
while on duty in the past month
74
59
... fired their service weapon at least once in their
career while on duty
30
24

Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

59
PEW RESEARCH CENTER

an arrest in the past month.
At the same time, about three-quarters (74%) of those who say they have grown more callous also
say they were verbally abused by a community member in the past month compared with 59% of
other officers. Three-in-ten officers who say they have grown more callous also report firing their
service weapons sometime during their police careers. By contrast, 24% of other police officers say
this.

www.pewresearch.org

60
PEW RESEARCH CENTER

4. Police, fatal encounters and ensuing protests
Shortly after noon on Aug. 9, 2014, Michael Brown, an 18-year-old black man, was shot and killed
by a white police officer in Ferguson, Missouri. Since 2015, almost 500 blacks have been fatally
shot by police. 11 Their deaths and the disputed circumstances surrounding many of these incidents
have sparked widespread protests over police tactics and raised new questions about the
relationship between the black community and the police.
A large majority of officers view these protests with deep skepticism. Fully two-thirds of police
(68%) say the demonstrations are motivated to a great extent by long-standing bias against the
police. A similarly sized majority (67%) characterizes the deaths of blacks during encounters with
the police that prompted these demonstrations as isolated incidents and not signs of a broader
problem between police and the black community. Black and white officers have profoundly
different views on this issue: 57% of black officers but only 27% of their white colleagues say these
deadly incidents point to larger issues between blacks and
Officers say policing is
police.
At the same time, these fatal encounters and the public outcry
they have generated have affected police work in fundamental
ways. More than eight-in-ten officers say their job is harder
now as a result of these incidents. Three-in-four officers say
interactions between police and blacks in their community have
grown more tense. About as many (72%) also say their
colleagues are less willing now to stop and question people who
seem suspicious or to use force when it is appropriate to do so.
Roughly nine-in-ten (93%) also say officers in their
departments are now more concerned about their safety, an
assessment the overwhelming majority of officers held even
before the ambush slayings of five Dallas police officers on July
7, 2016, by a black man who reportedly sought to avenge recent
shootings of black men by police.
These incidents also have prompted many police departments
to examine their own use-of-force policies, the survey finds.
About half of officers (46%) report that their department has

11

Pew Research Center analysis of data collected by The Washington Post.

www.pewresearch.org

harder now as a result of
incidents
% of officers saying high-profile
incidents involving police and
blacks have made …

Source: Survey of law enforcement officers
conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

61
PEW RESEARCH CENTER

modified its use-of-force policies or procedures, including 68% of officers in agencies with 2,600
or more sworn personnel. About six-in-ten officers (59%) say their agency has taken steps to
improve relations between police and blacks.

Large majority of officers say fatal
encounters are isolated incidents
Two-thirds of police officers (67%) say the
highly publicized deaths of blacks during
encounters with the police are isolated
incidents, while 31% describe them as signs of
a broader problem. Moreover, the survey finds
that majorities of officers in virtually every
major demographic group share this view,
with one striking exception. A majority of
black officers (57%) say these deaths are
evidence of a broader problem between police
and blacks, a view held by only about a quarter
of all white (27%) and Hispanic (26%) officers.

Most white, Latino officers say fatal
encounters between blacks and police
are isolated incidents; majority of black
officers disagree
% of officers saying the deaths of blacks during
encounters with police in recent years are …
Isolated
incidents
All officers

Less dramatic differences arise between other
demographic groups. Overall, male officers are
more likely than female officers to see these
deadly encounters as isolated incidents (69%
vs. 60%, respectively), in large part because
disproportionately more black women say

31

67
0

Men

Whites

30

69

Women

Black female officers in particular are more
likely to say these incidents signal a more farreaching concern. Among sworn officers, 63%
of black women say this, compared with 54%
of black men. By contrast, roughly equal
proportions of white male officers (27%) and
white female officers (29%) say the same.
Among Hispanic officers, about a quarter of
men (26%) and 32% of women say the
incidents reflect a broader problem.

Signs of a
broader problem

39

60

27

72

Blacks

57

43

Hispanics

72

26

Ages 18-34

72

27

35-44

70

29

45-49

64

35

50+

63

36

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

62
PEW RESEARCH CENTER

these incidents point to a more pervasive problem.
Younger police also are more likely than older officers to say these fatal police-black encounters
are isolated incidents, a view shared by 72% of officers younger than 35 but 63% of officers 50 and
older. Similarly, larger shares of rank-and-file officers and sergeants view these deaths as isolated
incidents (68% and 69%, respectively), compared with 61% of administrators.

Most officers say anti-police bias motivates protests
Overall, a majority of officers are deeply skeptical of the motives behind those who are protesting
high-profile deaths of blacks at the hands of police. However, black officers are far more likely
than their white colleagues to believe that protests are motivated by a desire to hold officers
accountable.
Overall, about nine-in-ten officers (92%) say long-standing anti-police bias is a motive for the
protests, comprising 68% who say it is a great deal of the motivation and about a quarter (24%)
who believe bias plays some role. Only 7% say
Most officers say protests mainly
bias against the police is not much or not at all
motivated by bias toward police
a reason for the protests.
At the same time, only about a third (35%) say
protesters felt a genuine desire to hold officers
accountable for their actions, including 10%
who say a great deal of the motivation for the
protests arises from the desire for
accountability. By contrast, about two-thirds
say the desire for accountability was not much
(36%) or not at all (28%) a motive for the
protesters’ actions.
Overall, at least nine-in-ten officers across
racial/ethnic groups – 95% of whites, 91% of
blacks and 90% of Hispanics – say the protests
are motivated at least to some extent by antipolice bias. But white officers are more likely
than black officers to say protests are driven a
great deal by bias against the police (72% vs.
59%, respectively). About two-thirds of

% of officers saying protests over deaths of blacks who
died during encounters with the police are motivated __
by …
A great deal

Some

Not much

Not at all

Long-standing bias against the police
NET 92%
68

24

42

A genuine desire to hold officers accountable for their
actions
NET 35%
10

25

36

28

Note: No answer category not shown. NETs calculated before
rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

63
PEW RESEARCH CENTER

Hispanic officers (65%) share this view,
placing them squarely between whites and
blacks.
The racial and ethnic divide among officers
opens even wider when they are asked the
degree to which protesters are motivated by a
genuine desire to hold police accountable for
their actions. A substantial majority of black
officers (69%) say the desire for accountability
was a great deal (34%) or some (35%) of the
impetus for the protests.

Black and white officers disagree over
extent to which protests are motivated
by desire to hold police accountable
% of officers saying protests over deaths of blacks who
died during encounters with the police have been
motivated a great deal or some by genuine desire to hold
police accountable
All officers

35

Men

34

Women
Whites

By contrast, about a quarter of white officers
(27%) say this, including only 5% who say it
had a great deal to do with the
demonstrations. Again, Hispanic officers
(42%) fall between whites and blacks on this
question.
Differences also emerge when the focus shifts
to gender, age and officer’s rank. Among all
sworn officers, women are significantly more
likely than men to say the protests are, at least
to some extent, efforts to produce
accountability (44% vs. 34%).

44
27

Blacks

69

Hispanics
Ages 18-34
35-44
45-49

42
30
33
37

50+
Rank-and-file
R&F officers
officers
Sergeants
Administrators

44
34
36
46

Note: Whites and blacks include only non-Hispanics. Hispanics are
of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”

Younger and older officers also hold
PEW RESEARCH CENTER
significantly different views. According to the
survey, three-in-ten officers younger than 35
say the desire for accountability is motivating the demonstrators a great deal or some. By contrast,
44% of officers 50 years old or older say this.
Police administrators are more sympathetic to the motives of protesters than lower-ranking
officers. Fully 46% of administrators say demonstrators are motivated a great deal or some by the
desire to make police accountable for their actions. By contrast only about a third of rank-and-file
officers (34%) and sergeants (36%) share this view.

www.pewresearch.org

64
PEW RESEARCH CENTER

Officers say high-profile incidents have made policing harder
Police work is a hard job that most officers say has become harder as a result of deaths of blacks
who died during encounters with police. Across every major demographic group analyzed for this
survey, about eight-in-ten officers or more say these high-profile incidents have made policing
more challenging and more dangerous.
Overall, fully 86% of officers say their job is harder now as a result of these deadly encounters.
Even officers from smaller departments that
typically serve smaller communities say it’s
Perceived tensions between blacks,
harder to be a police officer now; 84% of police
police have grown; officers more
in departments with fewer than 300 sworn
worried about safety and more reluctant
officers say their job is more difficult now –
to use appropriate force
and so do 89% of officers in big-city
% of officers saying each has happened in their
departments with 2,600 or more police.
department as a result of high-profile incidents
involving blacks and the police

An officer’s race matters, though not as much
on this question as in others in this survey.
Roughly nine-in-ten white officers (89%) say
policing is now harder, compared with 81% of
black officers. Among all sworn officers, black
male officers in particular are significantly less
likely than either white men or women to say
their job is harder now (79% for black men vs.
89% and 90% for white men and women,
respectively). Among black female officers,
84% say their job is more difficult now.

Officers have become more concerned about their safety
93
Officers have been more reluctant to use force when it is
appropriate
76
Interactions between police and blacks have become
more tense
75
Officers have become less willing to stop and question
people who seem suspicious
72

Officers say policing has changed
To measure the impact of fatal police-black
encounters on police work, the survey asked
officers if their department has changed in six
possible ways as a result of these high-profile
incidents involving blacks and the police. The
results document a range of consequences in
departments across the country.

The department has taken steps to improve relations
between police and blacks
59
The department has modified its policies or procedures
about the use of force
46
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

65
PEW RESEARCH CENTER



Increased safety concerns: Fully nine-in-ten officers (93%) say police in their department
have become more concerned about their safety. An analysis of survey responses taken before
and after five police officers died in an ambush-style attack by a gunman in Dallas suggests
that safety worries were high even before this incident: About nine-in-ten (92%) officers
expressed concerns about their physical safety before the attack, compared with 97% after the
attack, a small but significant increase.



Increased reluctance to use force: As a result of high-profile fatal police-black encounters,
three-quarters of officers (76%) say officers in their department have been more reluctant to
use force when it is appropriate. White and Hispanic officers (76% and 81%, respectively) are
more likely than black officers (68%) to say their colleagues are holding back in using more
forceful methods, even when such tactics are suitable for the situation.



Increased tension between police and blacks: Three-quarters of all officers report
increased tension between blacks and police in their
community. Female officers are somewhat more likely than
males to say this (80% vs. 75%).



Increased reluctance to stop and question suspicious
people: About seven-in-ten officers (72%) say the widely
publicized deaths of blacks who died during encounters
with police have made officers in their department less
willing to stop and question people who seem suspicious.
This finding raises the possibility that many officers are
responding to these incidents by “de-policing” – that is, by
not fully carrying out their law enforcement responsibilities
out of fear of becoming involved in a high-profile incident. 12
A larger share of white police officers (73%) than black
officers (64%) say their colleagues are now more hesitant to
question suspicious people. In particular, white male
officers are significantly more likely to say this (74%) than
white female or black male officers (65% for both).

At the same time, the survey finds that many departments have
initiated or modified programs to address issues raised by the

Regardless of race,
officers say their
colleagues are now more
reluctant to stop and
question suspicious
people
% of officers saying that officers in
their department have become less
willing to stop and question people
who seem suspicious
Whites
Blacks
Hispanics

73
64
72

Note: Whites and blacks include only nonHispanics. Hispanics are of any race.
Source: Survey of law enforcement officers
conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

For a popular account of de-policing in several American cities, see Kaste, Martin. Jan. 8, 2015. “When Morale Dips. Some Cops Walk the
Beat – But Do the Minimum.” National Public Radio.
12

www.pewresearch.org

66
PEW RESEARCH CENTER

fatal encounters between blacks and police.


Efforts to improve black-police relations: About six-in-ten officers (59%) say their
department has taken steps to improve relations with the black community. This view varies
substantially by the rank of the officer. Roughly eight-in-ten department administrators (79%)
say their department has initiated outreach efforts to the black community. By contrast, only
56% of rank-and-file officers and 64% of sergeants express a similar view, suggesting that word
about such efforts may not be filtering
down the ranks.

 Changes to the department’s use-offorce policies and procedures: Roughly
half of all officers (46%) say their
departments have modified their use-offorce protocols as a consequence of fatal
black-police encounters. Black officers are
more likely than whites to say this change
has happened in their department (59% vs.
42% for whites).

High-profile incidents have
disproportionately affected policing in
larger departments
% of officers in __ saying each has happened in their
department as a result of high-profile incidents
involving blacks and the police
Small departments (fewer than 300 officers)
Large departments (2,600 or more)
Officers have become less willing to stop and question
people who seem suspicious
54

Larger departments most affected
The series of high-profile deadly incidents
involving police and blacks has had the
greatest impact on officers in the country’s
largest police departments.
On every measure tested, police working in
departments with 2,600 sworn officers or
more are significantly more likely to say their
department has been affected by these
incidents than those in police agencies with
fewer than 300 officers. Among departments
with more than 100 sworn police, three-in-ten
officers nationally work in these large
departments, while a slightly smaller share are
employed in agencies with fewer than 300

86
Interactions between police and blacks have become
more tense
61
87
Officers have become more reluctant to use force when it is
appropriate
63
85
Officers have become more concerned about their safety
88
95

Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

67
PEW RESEARCH CENTER

officers (24%). The others work in departments with 300 to 2,599 officers (45%).
About nine-in-ten officers (86%) in large departments with at least 2,600 officers say their
colleagues have become less willing to stop and question suspicious individuals. By contrast, about
half (54%) in small departments with fewer than 300 officers say this.
Fully 87% of officers in large departments, but 61% in small agencies, say relations with blacks in
their community have grown more tense as a consequence of recent black deaths during
encounters with the police. Some 85% of officers in large police departments say their colleagues
are more reluctant now to use force even when it is appropriate to do so. By contrast, about six-inten (63%) in small departments say this.
On just one measure are there only modest differences between police in larger and smaller
departments: Fully 95% of officers in the largest departments say police are now more concerned
about their safety, a view shared by 88% of
those in small agencies.
Large departments more likely to have
Equally striking differences between larger
and smaller departments emerge when officers
were asked if their department has taken steps
to improve relations between police and the
black community.
About two-thirds of officers in large
departments (66%) say their department has
attempted to improve relations between
officers and the black community as a result of
fatal police-black encounters, while 35% of the
officers in small departments say the same.

moved to modify use-of-force policies or
procedures
% of officers in ___ saying each has happened in their
department as a result of high-profile incidents
involving blacks and the police
Small departments (fewer than 300 officers)
Large departments (2,600 or more)
The department has taken steps to improve
relations between police and blacks
35
66
The department has modified its policies or
procedures about the use of force
19

Similarly, about two-thirds of officers in larger
departments (68%) say their department has
modified its use-of-force rules in response to
fatal police-black encounters, while 19% of the
officers in small departments say the same.

68
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

68
PEW RESEARCH CENTER

5. Reimagining the police through training and reforms
Recent fatalities of blacks during police
encounters have brought police training and
reforms to the forefront of conversations on
how to prevent the use of unnecessary force.
Recommendations have been made to prevent
these types of situations from occurring, such
as requiring officers to wear body cameras and
training officers on how to de-escalate
situations to reduce the need to use force. 13
The survey finds that a majority of officers
favor the use of body cameras by police
officers. About half think that wearing body
cameras will make police more likely to act
appropriately. But just one-third of officers
think that it would make the public more likely
to cooperate with the police.

Roughly two-thirds of officers say they
favor the use of body cameras
% of officers saying they personally ___ the use of body
cameras, regardless of whether their department uses
them
Oppose

Favor

All officers

33

66

Rank-and-file
officers

34

65

Sergeants
Administrators

31

67
18

82

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”

Despite the national attention given to training
and reforms aimed at preventing the use of
PEW RESEARCH CENTER
unnecessary force, relatively few (half or less
than half of rank-and-file officers) report
having had at least four hours of training in some key areas over the past 12 months. For example,
half of rank-and-file officers say, over the past year, they have had at least four hours of training in
nonlethal methods to control a combative or threatening individual, and 46% say they have had at
least four hours of training on how to deal with individuals who are having a mental health crisis.
The survey also asked officers their views on how useful several different approaches of policing
are today. A majority of police say that requiring officers to show respect, concern and fairness
when dealing with the public is very useful. But less than half of those who view this as a very or
somewhat useful approach say that their department’s leadership provides a great deal of support
to officers who want to do so. A narrow majority of officers say that patrolling high-crime areas is
very useful in policing today. About three-in-ten officers (32%) who view this as a useful strategy
say their leadership gives a great deal of support to officers who want to take this approach.
In 2016, the Police Executive Research Forum, a nonprofit focused on critical issues in policing, developed 30 guiding principles on the use
of force that includes recommendations on policies, training, tactics and equipment.
13

www.pewresearch.org

69
PEW RESEARCH CENTER

Most officers favor use of body
cameras, but many are skeptical
that they would change behavior
A majority (66%) of officers favor the use of
body cameras by police. Administrators are
particularly likely to favor their use: 82% of
administrators say they favor the use of body
cameras, compared with about two-thirds of
sergeants (67%) and rank-and-file officers
(65%).
Half of officers say that wearing body cameras
would make officers more likely to act
appropriately when dealing with the public.
But a similar share (44%) say that wearing a
body camera would make no difference in the
way officers interact with the public. Only 5%
say wearing a body camera would make
officers less likely to act appropriately.
Black officers are considerably more likely
than their white and Hispanic counterparts to
say wearing body cameras would make officers
act more appropriately. Roughly seven-in-ten
(71%) black officers say this, compared with
about half of white (46%) and Hispanic (53%)
officers.
Officers with higher ranks and more years of
experience are also particularly likely to say
that wearing body cameras would make
officers more likely to act appropriately when
dealing with the public. About seven-in-ten
(69%) administrators say this, compared with
roughly half of sergeants (53%) and rank-andfile officers (48%). And, while 57% officers
with 20 or more years of experience say this,

Black officers more likely than whites,
Hispanics to say body cams would
change police behavior
% of officers saying that wearing body cameras would
make officers ____ when dealing with the public
More likely to
act appropriately

It would make
no difference

All officers

Whites

Less likely to
act appropriately
44

5

49

5

50

46

Blacks
Hispanics

5

25

71
42

53

4

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Administrators see more benefit to body
cams than sergeants, rank-and-file
officers
% of officers saying that body cameras on officers would
make members of the public …
More likely
to cooperate

It would make
no difference

Less likely
to cooperate

All officers

33

56

10

Rank-and-file
Rank-and-file
officers
officers

32

58

10

Sergeants

34

55

11

Administrators

52

41

7

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

70
PEW RESEARCH CENTER

only 49% of those with 10 to 19 years of experience and 43% of officers with less than 10 years of
experience agree. Among rank-and-file officers, these differences by years of experience remain.
When asked about the effect of body cameras on the way the public interacts with police officers,
only one-third of officers say that body cameras would make the public more likely to cooperate.
Just over half (56%) say wearing body cameras would not make a difference in the way the public
interacts with police officers. One-in-ten say that wearing body cameras would make members of
the public less likely to cooperate with officers.
Views on this differ significantly by rank: 52% of administrators say that wearing body cameras
would make the public more likely to cooperate with officers. Smaller shares of sergeants (34%)
and rank-and-file officers (32%) hold this view.
Black officers are also more
likely than white officers and
Hispanic officers to say that
wearing body cameras would
make members of the public
more likely to cooperate.
Some 44% of black officers
say this, compared with 31%
of white officers and 33% of
Hispanic officers.

About one-in-four rank-and-file officers say they have
had no training on how to de-escalate a situation in
the past 12 months
% of rank-and-file officers saying they have received ___ training in each
of the following areas in the past 12 months
4 or more hours

Less than 4 hours

Firearms training involving shootdon't shoot scenarios

53

Most officers report
having some use-offorce training in past
12 months

Nonlethal methods to control a
combative or threatening individual

50

How to deal with individuals who
are having a mental health crisis

46

The survey asked police
officers how much training
they had in several key areas,
which has been
recommended to protect the
public and police from
encounters where
unnecessary force is used.

How to de-escalate a situation so it
is not necessary to use force

44

Bias and fairness
How to deal with people so they feel
they've been treated fairly and
respectfully

39

37

None
31

31

34

32

37

34

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

15

18

19

24

23

28

71
PEW RESEARCH CENTER

About half of rank-and-file officers – those who are assigned to a beat where they routinely
interact with the public – say they have had at least four hours of firearms training involving
shoot-don’t shoot scenarios (53%) and nonlethal methods to control a combative or threatening
individual (50%) in the past 12 months.
Some 46% of rank-and-file officers have had at least four hours of training in how to deal with
individuals who are having a mental health crisis, and 44% say they have had at least four hours of
training in how to de-escalate a situation so it is not necessary to use force. About four-in-ten
rank-and-file officers say they have received at least four hours of training in bias and fairness
(39%) and how to deal with people so they feel they’ve been treated fairly and respectfully (37%).
Rank-and-file officers with less than five years of experience are more likely than those with more
experience to say they have had these types of trainings in the past 12 months. For example, 63%
of rank-and-file officers with less than five years of experience say they have had at least four
hours of training in nonlethal methods to control a combative or threatening individual, compared
with 47% of rank-and-file officers with five or more years of experience.

Nearly four-in-ten officers say it’s
very important to have knowledge
of policing strategies scientifically
shown to be effective
Some 36% of police officers say it is very
important to have a good knowledge of what
scientific research shows to be effective
policing strategies. But high-ranking
administrators are more likely than rank-andfile officers or sergeants to say this is the case.
About half of administrators (49%) say it is
very important, while 35% of rank-and-file
officers and 34% of sergeants say the same.
There are also differences along racial and
ethnic lines. Some 44% of black officers and
46% of Hispanic officers say it is very
important to have a good knowledge of
scientific research on effective policing,

Administrators place more importance
on the science of policing
% of officers saying it is ____ for law enforcement
officers today to have a good knowledge of what
scientific research shows are effective policing strategies
Very
important

Somewhat
important

Not too
important

Not at all
important

All officers

36

49

12

3

Rank-and-file
Rank-and-file
officers
officers

35

50

13

3

Sergeants

34

51

13

2

Administrators

49

43

7 1

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

72
PEW RESEARCH CENTER

compared with 32% of white officers.
New officers (43%) are more likely than officers with five or more years of experience (35%) to say
it is very important to have good knowledge of what scientific research shows to be effective
policing strategies.

A majority of officers say showing respect, concern and fairness when
dealing with the public is very useful in policing today
Some police reform efforts are aimed at more subtle shifts in how the police interact with
community members and where they focus their resources.
A majority (65%) of officers say that today in policing it is very useful for departments to require
officers to show respect, concern and fairness when dealing with the public – an approach referred
to as procedural justice. Among those who view this as a very or somewhat useful strategy, 45%
say that their department’s leadership gives a great deal of support to officers who want to do this.
An additional 42% say their leadership gives a fair amount of support.
About six-in-ten (58%) officers say that requiring officers to patrol more frequently in high-crime
areas is very useful. This approach is called “hot spot” policing, because it concentrates policing
efforts in the small areas
where crime is concentrated.
Most officers see value in showing respect and
Among officers who view this
fairness when dealing with the public
as a useful approach, 32% say
% of officers saying that today in policing it is ___ to require officers to do
that their department’s
each of the following
leadership gives officers who
Very
Somewhat
want to patrol more
useful
useful
Show respect, concern and
frequently in high-crime
fairness when they deal with
65
30
areas a great deal of support.
the public
An additional 46% say the
department’s leadership gives
Patrol more frequently in
58
34
high-crime areas
a fair amount of support.
Fewer officers view the
community policing
approach as very useful
today: About four-in-ten
(41%) say that requiring

Be responsive to community
concerns and work in close
partnership with the
community to solve problems

41

Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

47

73
PEW RESEARCH CENTER

officers to be responsive to community concerns and work in close partnership with the
community to solve problems is very useful. Among officers who view this as a useful strategy,
roughly three-in-ten (32%) officers say their department’s leadership gives a great deal of support
to officers who want to do so. An additional 50% say their leadership gives a fair amount of
support.

Most officers say responding effectively to mental health crises is an
important role of police officers
About three-quarters (76%) of officers say that
responding effectively to people who are
having a mental health crisis is an important
role of police officers, and an additional 12%
say that it is a role of police officers but not an
important one. About one-in-ten (11%) officers
say this is not a role of police officers.

Most officers say responding effectively
to mental health crises is an important
role of police officers
% of officers saying that they think responding
effectively to people who are having a mental health
crisis is ___ of police officers
An important role

76

While majorities of officers across all
A role, but not an
12
important one
demographic groups studied and agency
characteristics say responding to mental
Not a role
11
health crises is an important role of police
officers, those in larger department are less
Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
likely than officers in smaller department to
14, 2016.
“Behind the Badge”
say this is the case. About seven-in-ten (68%)
PEW RESEARCH CENTER
officers in departments with at least 2,600
officers say responding effectively to people
who are having a mental health crisis is an important role of police officers, compared with 83%
among those in departments with fewer than 500 officers and 78% in departments with 500 to
2,599 officers.

About half of officers say local police should take an active role in
identifying undocumented immigrants
Officers are slightly more likely to say that local police should take an active role (52%) in
identifying undocumented immigrants rather than leaving this task mainly to federal authorities
(46%).

www.pewresearch.org

74
PEW RESEARCH CENTER

White officers and those with more years of
experience are more likely than their
counterparts to say that local police should
take an active role in identifying
undocumented immigrants. About six-in-ten
(59%) white officers say this, compared with
35% of black officers and 38% of Hispanic
officers. And about six-in-ten (58%) officers
with at least 20 years of experience say this,
compared with 52% of those with 10 to 19
years of experience and 46% of those with less
than 10 years of experience.

About half of officers say local police
should take the lead in identifying
undocumented immigrants
% of officers saying that when it comes to identifying
undocumented or illegal immigrants …
It should be left
mainly to federal
authorities
All officers

Hispanics

52

46

Whites
Blacks

Local police
should take
an active role

59

41
64
60

35
38

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

75
PEW RESEARCH CENTER

6. Police views, public views
Police and the public hold sharply different views about key aspects of policing as well as on some
major policy issues facing the country. For example, most police say more officers are needed to
adequately patrol their communities, while the majority of the public doesn’t think more officers
are necessary. A majority of officers oppose a ban on assault-style weapons, while a majority of the
public favors a ban on these weapons. More than eight-in-ten police say people don’t understand
the risks and rewards of police work well, while an equally large majority of the public says they
do.
At the same time, there are areas of broad agreement between officers and the public. Majorities of
the police and public favor the use of body cameras by officers to record interactions with the
public. Large majorities of police and the public also support easing some legal restrictions on
marijuana, though the public is more likely than officers to support the legalization of marijuana
for both personal and medical use (49% vs.
32%).
These contrasting views and striking
similarities emerge from two surveys, one of
7,917 sworn police officers conducted online
May 19-Aug. 14, 2016, and the other a
nationally representative survey of 4,538
adults conducted Aug. 16-Sept. 12, 2016, by
mail and online. The surveys included a
number of identically worded questions, which
allowed for direct comparisons of how officers
and the public see the role of the police in their
communities and how they view recent deaths
of blacks during encounters with police, as well
as to capture their views on some major policy
issues, including gun control, the use of body
cameras by officers and assessments of racial
progress.
Some of the sharpest differences between the
police and the public emerge over views on
deaths of blacks during encounters with police
in recent years and the protests that many of

Police, public differ on perceptions of
deadly black-police encounters
% saying the deaths of blacks during encounters with
police in recent years are …
Signs of a
broader problem

Isolated
incidents
All officers

67

All public

31
39

60

Among whites
Officers

72

Public

27
44

54

Among blacks
Officers
Public

43

57
18

79

Note: No answer category not shown. Whites and blacks include
only non-Hispanics.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

76
PEW RESEARCH CENTER

those incidents ignited. For example, 67% of the police but only 39% of the public describe these
deadly encounters as isolated incidents rather than signs of a broader problem between blacks and
police. When this overall finding is analyzed by race, an equally striking result snaps into focus:
About seven-in-ten white officers (72%) but fewer than half of all black officers see these
encounters as isolated incidents. By contrast, majorities of black officers (57%) as well as the
public overall (60%) say the incidents are signs of a broader problem between police and the black
community.
When the subject shifts to overall views on racial progress, large differences again emerge between
the public and the police and also between blacks and whites within each group. For example,
when police and the public are asked if the country has made the changes needed to give blacks
equal rights with whites, fully eight-in-ten police officers – including 92% of white officers but only
29% of black officers – say the necessary changes have been made. By contrast, about half (48%)
of the public, including 57% of whites but only 12% of blacks, says the country has made the
changes needed for blacks to have equal rights with whites.
The remainder of this chapter examines these and other related findings in greater detail. The first
sections compare and contrast police and public views on the
role of police in the community and how each group views the
Protectors or enforcers?
risks and rewards of police work. The next sections describe
Majority of officers, half
how each group views recent deadly encounters between police
of public say police are
and blacks and also examines police and public attitudes
both
toward the protests that followed many of these incidents. The
% of officers and the public saying
final section examines police and public attitudes on some
they see themselves/see their local
police more as …
current issues relevant to law enforcement, including gun
Officers
Public
policy, legalization of marijuana and racial equality.

How the public sees the police, how police see
themselves
Protectors, enforcers or both – what do Americans see when
they look at their local police? And do their perceptions of the
police align with what officers say is their primary role?
Overall, about six-in-ten (62%) officers say their primary role is
to serve as both protectors and enforcers; among the public,
about half (53%) view their local police this way.

www.pewresearch.org

Protectors
Enforcers

16
8

Both
Both
equally
equally

31

29
62
53

Note: No answer category not shown.
Source: Survey of law enforcement officers
conducted May 19-Aug. 14, 2016; survey
of U.S. adults conducted Aug. 16-Sept. 12,
2016.
“Behind the Badge”
PEW RESEARCH CENTER

77
PEW RESEARCH CENTER

At the same time, three-in-ten officers (31%)
say their primary role is to serve as protectors,
about twice the share of the public (16%) who
see their local police in that way.
An even larger disparity between police-public
views emerges over the enforcement role of
police. Only 8% of officers say they mainly see
themselves as enforcers – the long arm of the
law – yet fully three times the share of the
public (29%) see their local police that way.

Majority of public wants no change in
size of force; officers say more police
needed
% of public saying they would prefer __ than currently
exists in their local area
A larger
police presence
34

A smaller
police presence
8

No change

57

% of officers saying their department __ have enough
officers to adequately police the community
Does not

This disparity over how the public views police
and how officers see their role is partially
explained by race. Blacks are significantly
more likely than whites to see their local police
as mainly enforcers (39% vs. 26%) and less
likely to see officers as both protectors and
enforcers (43% vs. 57%).
Overall, 46% of Hispanic adults see police in
their community as both enforcers and
protectors, while 33% view them as enforcers
and 14% as protectors. Among Hispanic
officers, about two-thirds (65%) see their role
to be both protectors and enforcers, while 7%
say they are enforcers and 28% consider
themselves to be protectors.

Contrasting views on size of police force
When it comes to manpower, police are
unequivocal: More than eight-in-ten officers
(86%) say their department does not have
enough police to adequately patrol their
community. By contrast, a majority of the
public (57%) wants no change in the size of the
local police force. About a third of the public

Does

86

13

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Among blacks, police and public have
different views of role of police
% saying they see themselves/see their local police
more as …
Protectors

Enforcers

Both equally

Among whites
Officers
Public

9

32
17

59
57

26

Among blacks
27

Officers
Public

69

4

43

39

14

0
Among Hispanics
28

Officers
Public

14

7
33

65
46

Note: No answer category not shown. Whites and blacks include
only non-Hispanics. Hispanics are of any race.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

78
PEW RESEARCH CENTER

(34%) want more officers in their local area,
and 8% favor fewer officers.
Among the public, these differences are linked,
in part, to how they see their local police.
Among those who view the local police as
mainly being enforcers, a quarter say they
want more officers and 19% would favor a
smaller police department. The remaining
54% favor no change.
At the same time, roughly a third (36%) of
those who see the police as both protectors
and enforcers would prefer to see more
officers. Only 4% favor a smaller force, while
59% prefer the current level of policing.
Similarly, about a third of those who view their
police as protectors (30%) favor a larger police
presence, 11% would like a smaller force and
59% prefer no change. (The views of police on
whether there are enough officers in their
communities are far more unequivocal: About
eight-in-ten or more in each group says their
department falls short of having the number of
officers their community needs.)

Americans who see local police as
enforcers less likely to want a greater
police presence
% saying they would like to see a ___ police force in
their local area, among the public who view local police
more as protectors/enforcers/both equally
Larger
Protectors
Enforcers
Both equally

30
25

11

59

19

54

36

4

59

PEW RESEARCH CENTER

Do Americans understand the
challenges police face? Public says yes,
police say no
% of public saying they understand the risks and
challenges that police face on the job …

Police work: Great risks, great
frustrations

Somewhat
well
NET 83%
38

Not too
well

Not well
at all
45

13

3

% of officers saying the public understands the risks
and challenges that police face on the job …
NET 14%
1

Fully eight-in-ten Americans (83%) say they
understand the risks and challenges of police
work – including 38% who believe they
understand the risks very well. By contrast,

No change

Note: No answer category not shown.
Source: Survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”

Very
well

The overwhelming majority of Americans say
they understand the risks and challenges that
police face. And an equally lopsided share of
police disagrees.

Smaller

12

46

40

Note: No answer category not shown. NETs calculated before
rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

79
PEW RESEARCH CENTER

fully 86% of the police say the public does not fully comprehend the trials that officers face –
including 40% who say Americans don’t understand well at all the risks and challenges of police
work.
Another survey finding provides a striking example of an apparent disconnect between what the
public thinks police work is like and the reality of law enforcement.
Perhaps influenced by popular television police dramas that routinely feature vividly
choreographed shootouts, more than eight-in-ten Americans (83%) believe that typical police
officers fire their service weapon while on duty at least once in their career – and about three-inten (31%) believe police discharge their weapon at least a few times a year.
In fact, only about a quarter of all officers (27%) say they have ever fired their service weapon. 14

Police work more dangerous, more frustrating
But how do police view the risks and rewards of their work, and how do those views differ from
Americans in other occupations? To partially answer those questions, the surveys asked officers
and employed Americans how often they
worried about their physical safety while at
Officers more worried about their
work, how often their job made them feel
personal safety on the job
frustrated and how often it made them feel
% saying they __ have serious concerns about their
physical safety when they are at work
fulfilled.
Average police officers are three times as likely
as workers overall to say they nearly always or
often have serious concerns about their
physical safety while on the job (42% vs. 14%).
Employed Americans, meanwhile, are about
four times as likely as officers on average to
say they hardly ever or never seriously worry
about their physical well-being at work (67%
vs. 16%).

Nearly always/
Often
Officers

All employed

42

14

19

Hardly ever/
Sometimes Never
42

16

67

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Officers also are more likely on average than
employed Americans overall to say their jobs frequently make them feel frustrated and somewhat
The survey questions specifically asked the public and officers not to include in their count or estimate instances where officers fired their
service weapon “on a gun range or while training.”

14

www.pewresearch.org

80
PEW RESEARCH CENTER

less likely to feel fulfilled by their work. Half
(51%) of officers say their job nearly always or
often frustrates them, compared with 29% of
all workers. A larger share of white officers
report feeling frustrated by their job than do
white workers overall (54% vs. 28%).
At the same time, about four-in-ten officers
(42%) say their work frequently makes them
feel fulfilled, compared with half of employed
adults (52%) who feel that way. There was no
significant difference between white and black
officers.

Officers, public agree that police
work is more difficult now
The public and officers agree that recent
deaths of blacks during incidents with police
and the protests they have sparked have added
to the challenges of police work. More than
eight-in-ten officers (86%) say their job is
harder now as a result of the protests. At the
same time, seven-in-ten Americans believe
police work has become more dangerous in the
past five years.
On both measures, solid majorities of whites
and blacks agree police work is harder now,
though whites are more likely than blacks to
say policing has become more challenging.
About nine-in-ten white officers (89%) and
81% of black police say their job has gotten
harder. Similarly, roughly three-quarters of
whites (74%) in the general population and
60% of blacks say policing has become more
hazardous in recent years.

Police work: More frustrating, less
fulfilling than most other jobs
% saying their work nearly always or often makes them
feel …
Officers

All employed
51

Frustrated

29
42

Fulfilled

52

Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Officers say policing is harder now,
public views it as more dangerous
% of officers saying high-profile incidents involving
police and blacks have made their job …
Harder

Easier*

No difference

86

12

*Less than 0.5%

% of public saying policing is __ compared with five
years ago
More dangerous
70

Less dangerous

About the same
6

24

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

81
PEW RESEARCH CENTER

Public, police see deadly policeblack encounters differently
Police and the public describe the recent fatal
incidents involving blacks and the police in
very different ways.
Roughly two-thirds of the police (67%) say
these deadly encounters are isolated incidents,
while about three-in-ten (31%) say they are
signs of serious problems between law
enforcement and the black community. But
when the public is asked to consider these
incidents, the result is virtually reversed: Sixin-ten say these encounters are signs of a
broader problem, while 39% describe them as
isolated incidents.
These differences grow sharper when race is
added to the analysis. Black and white officers
see these incidents very differently, as do
whites and blacks in the general public.
For example, about seven-in-ten white police
officers (72%) and 44% of whites overall say
fatal black-police encounters are isolated
incidents. By contrast, only about four-in-ten
black officers (43%) and 18% of blacks overall
share this view.
At the same time, black officers are about
twice as likely as blacks nationally to describe
these encounters as isolated incidents (43% vs.
18%). And while a narrow majority of black
officers (57%) say the incidents are signs of a
broader problem, a much larger majority of
blacks overall (79%) express this view. In fact,
roughly similar shares of black police and

Majority of police say fatal police-black
encounters are isolated incidents;
majority of the public says the
encounters point to a bigger problem
% saying the deaths of blacks during encounters with
police in recent years are …
Isolated
incidents
Officers

67

Signs of a broader
problem
31

Public

39

60

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Large racial differences on perceptions
of deadly black-police encounters
among police, the public
% saying the deaths of blacks during encounters with
police in recent years are …
Isolated incidents

Signs of a broader problem

Among whites
Officers

72

Public

27
44

54

Among blacks
Officers
Public

43

57
18

79

Note: No answer category not shown. Whites and blacks include
only non-Hispanics.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

82
PEW RESEARCH CENTER

whites overall – 57% and 54%, respectively –
see these incidents as pointing to larger issues
between blacks and law enforcement.

Anti-police bias is seen as a motive
for protests by most officers and
public

Most officers, public agree anti-police
bias is a motive for protests
% saying protests over deaths of blacks who died during
encounters with the police are motivated __ by longstanding bias against the police
A great deal

Some

Public

Not at all

68

Officers

Large majorities of the police and the public
agree that long-standing anti-police bias was
at least some of the principle behind the
protests that have followed many of the recent
fatal incidents involving blacks and the police.

Not much

24

41

38

4 2

12

7

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

About nine-in-ten officers (92%) say the
protests were motivated by bias toward the
police, including 68% who say this was a great
deal of the reason behind the demonstrations.
A smaller but still substantial 79% majority of
the public agrees that prejudice against the
police provided at least some of the impetus
for the protests, including 41% who see this as
a major motivation.
A familiar pattern emerges when race is
factored into the analysis. Black and white
officers differ somewhat about the motives of
the protesters, and the views of each group
differ with those of all blacks and whites.
Fully nine-in-ten white officers (95%) and 85%
of whites nationally say the protests are
motivated at least somewhat by anti-police
bias. Underlying this modest difference is the
much larger share of white officers who feel
that long-standing animosity toward police is a

Many blacks and whites say anti-police
bias is a protest motive
% saying protests over deaths of blacks who died during
encounters with the police are motivated __ by longstanding bias against the police
A great deal

Some

Not much

Among whites

Not at all

NET 95%

Officers

23

72

3 2

NET 85%
Public

47

Among blacks

38

10 5

NET 91%
59

Officers

32

63

NET 56%
Public

25

32

21

20

Note: No answer category not shown. NETs calculated before
rounding. Whites and blacks include only non-Hispanics.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

83
PEW RESEARCH CENTER

great deal of the protesters’ motivation (72% of
white officers vs. 47% for all whites).

Police, public disagree about
accountability as a protest motive

Among blacks, the disparity between police
and the public is even greater than it is among
whites. About nine-in-ten black officers (91%)
say anti-police feelings are a reason for the
protests. By contrast, 56% of blacks overall
share this view.

% saying protests over deaths of blacks who died during
encounters with the police are motivated ___ by the
genuine desire to hold police accountable

Again, the belief that anti-police bias is a
major reason behind the demonstrations is
more strongly held by black officers. Black
officers are more than twice as likely as blacks
generally to say bias was a great deal of the
reason for the demonstrations (59% of black
officers vs. 25% of all blacks).

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”

White officers more skeptical that
accountability motivated protests
The police and the public also disagree about
how important a motivation the desire to hold
police accountable was to protesters. The
difference is dramatic: Only about a third of all
officers (35%) say the desire to make officers
answerable was at least some of the motivation
for the demonstrations, while 65% of the
public says accountability was a factor.
A different pattern emerges when blacks and
whites are asked the degree to which they
believe the protests are genuine attempts to
force police accountability. White officers
stand apart; they are far less likely than whites
generally, black officers or blacks to see

Not much/Not at all
Officers

A great deal/Some

64

35

Public

65

34

PEW RESEARCH CENTER

White officers less likely than other
groups to say accountability a motive
% saying protests over deaths of blacks who died during
encounters with the police are motivated ___ by a
genuine desire to hold officers accountable
A great deal

Some

Not much

Not at all

Among whites
NET 27%
Officers

5

40

22

32

NET 63%
Public
Among blacks
Officers

35

27

25

12

NET 69%
35

34

21

10

NET 79%
Public

55

24

11 7

Note: No answer category not shown. NETs calculated before
rounding. Whites and blacks include only non-Hispanics.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

84
PEW RESEARCH CENTER

holding officers answerable for their actions as a major goal of the protests.
About a quarter (27%) of white officers say accountability motivated the protests; by contrast,
more than twice the share of whites overall (63%) say this. In fact, a quarter of whites overall
(27%) say the desire for police accountability was a great deal of the reason for the protests –
identical to the share of police who say accountability was either a great deal (5%) or some (22%)
of the motivation, combined.
Blacks, both officers and in the public, see the desire for accountability as a driving factor behind
the protests. About seven-in-ten black police officers (69%) say concerns about police
accountability played at least some role in the protests, a view shared by 79% of all blacks.
Moreover, blacks nationally are significantly more likely than black officers to say this was a great
deal of the motivation for demonstrators (55% vs. 34%).

Broad support for body cameras
One consequence of recent fatal encounters between police and blacks has been the growing call
for police to wear video cameras to record interactions between officers and the public. While
some law enforcement organizations, including the police unions in Miami and Boston, have
attempted to slow down efforts to make officers wear “body cams,” the surveys find that a clear
majority of officers and a larger share of the public support their use.
Two-thirds of the police (66%) and an even
larger share of the public (93%) favor the use
of body cameras by police to record
interactions between officers and the public.
However, the surveys also find that police see
relatively fewer benefits than the public does
from the use of body cams by officers.
About six-in-ten Americans (59%) but only a
third of police say body cams would make
members of the public more likely to
cooperate with officers. By contrast, a majority
of police (56%) says body cams would make no
difference, a view shared by about a third
(35%) of the public. Only 5% of the public and

Majority of police and a larger share of
the public favor body cameras
% saying they __ the use of body cameras by police
Oppose
Officers

Favor

33

Public

66
6

93

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

85
PEW RESEARCH CENTER

10% of the police say members of the public
would be less likely to obey officers who are
wearing body cameras.

Public sees more benefits than police
from use of body cameras
% saying body cameras on officers would make …

Police are somewhat more convinced about
the positive effects of body cameras on police
behavior than on the public’s behavior. Half of
officers and two-thirds of the public (66%) say
a police officer would be more likely to act
appropriately when wearing a body cam. At
the same time, 44% of officers and 27% of the
public doubt that wearing body cams would
have an impact on police behavior, while small
shares of officers and the public say officers
would be less likely to act appropriately (5%
and 6%, respectively).

Broad support from police, public
for some gun law reforms
Police officers are considerably more likely
than the general public to say it is more
important to protect the rights of Americans to
own guns than it is to control gun ownership
(74% of officers vs. 53% of the public). At the
same time, there is widespread agreement
between police and the public on several key
gun law reforms. For example, more than
nine-in-ten officers and almost the same share
of the public favor laws that would prevent the
mentally ill from purchasing guns (95% and
87%, respectively). And about the same
proportions of the police and the public favor
background checks for people who buy
weapons at a gun show or from a private
individual (88% and 86%, respectively).

Public more likely
to cooperate
with officers
Officers

Public less likely
to cooperate

33

Public

10

Officers

56

59

Officers more
likely to act
appropriately

No difference

5

35

Officers less
likely to act
appropriately
50

Public

No difference

5
66

44
6

27

Note: No answer category not shown. Officers were shown both
questions on one screen. Half the sample of the public was asked
about officers acting more or less appropriately and the other half
was asked about members of the public being more or less likely to
cooperate with officers.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Police more supportive of gun rights
than public
% saying it is more important to …
Officers

Public

Protect the right
of Americans to
own guns
Control gun
ownership

74
53
25
46

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

86
PEW RESEARCH CENTER

A majority of police and a larger share of the
public also support the creation of a federal
database to track gun sales (61% and 71%,
respectively).
However, the consensus on guns vanishes
when the focus turns to assault-style weapons.
About two-thirds of Americans (64%) but only
about a third of police (32%) favor outlawing
assault weapons.
The gender gap among police on this issue is
among the largest of any question in this
survey: A majority of female officers (57%)
favor a ban on assault weapons, compared
with about a quarter of their male colleagues
(27%). This disparity mirrors the overall
gender gap in the country as a whole: 74% of
women and 54% of men favor making these
weapons illegal.

A majority of police, public
support easing some restrictions
on marijuana

Police, public agree on a range of new
gun control measures, disagree on
assault weapons ban
% saying they favor …
Officers

Public

Laws to prevent
mentally ill from
purchasing guns

95
87

Making private gun
sales and sales at
gun shows subject to
background checks

88
86

Creating a federal
database to track
all gun sales

A ban on assaultstyle weapons

61
71
32
64

Note: Officers were shown all four questions on one screen. Half the
sample of the public was asked about laws regarding the mentally ill
and background checks and the other half was asked about laws on
assault weapons and a federal database.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”

As more jurisdictions move to decriminalize or
legalize the private use of marijuana by adults,
PEW RESEARCH CENTER
large majorities of the police and the public
favor easing restrictions on the drug. However,
a larger share of the public than police favor legalization of marijuana for personal and medical
use (49% vs. 32%).

Overall, about seven-in-ten officers support allowing medical use of marijuana (37%) or favor the
legalization of the drug for both personal and medical use (32%). The public is more favorably
inclined than police toward relaxing marijuana laws; more than eight-in-ten Americans support
either legalizing marijuana (49%) or allowing only medical use of the drug (35%).

www.pewresearch.org

87
PEW RESEARCH CENTER

The surveys found little support among the
public for outlawing marijuana use under any
circumstances (15%). However, police are
twice as likely as all adults to favor an outright
ban on the drug (30%).
As with younger adults generally, officers
younger than 35 are more likely than those
ages 50 to 60 to favor permitting personal and
medical use of marijuana (37% vs. 27%).
Among the public, a majority of adults (63%)
under the age of 45 favor legalization.

Police say no more changes needed
to achieve racial equality; public
divided

A majority of police, public favor
relaxing marijuana laws
% saying marijuana use by adults …
Should be legal for medical and personal use
Should be legal for medical use only
Should not be legal
NET 68%
Officers

32

37

30

NET 84%
Public

49

35

15

Note: No answer category not shown. NETs are calculated before
rounding.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

The wide disparities in the views of blacks and
whites in American society over whether more
changes are needed to achieve racial equality
loom even larger in the country’s police
departments.
Overall, the surveys find that police are
significantly more likely than the public to say
the country has made the changes necessary to
give blacks equal rights with whites (80% vs.
48%). By contrast, half of the public believes
the country still needs to make changes to
achieve racial equality, a view shared by only
16% of police.

Younger police and younger adults more
likely to favor legalization of marijuana
% saying marijuana should be legal for medical and
personal use by adults
Officers

37

Ages 18-34

67
32

35-44

57
32

45-49
50-60

Underlying these overall results are sharp
disagreements between blacks and whites on
this issue – a racial divide that is wider within
America’s police departments than it is in the
country as a whole.

Public

47
27
45

Note: Less than 2% of the police sample was older than 60 so the
analysis was restricted to those in the police and general public
samples who were 60 or younger.
Source: Survey of law enforcement officers conducted May 19-Aug.
14, 2016; survey of U.S. adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

88
PEW RESEARCH CENTER

Fully nine-in-ten white police
officers (92%) say the country
has made the needed changes
to achieve racial equality.
Nationally, a modest 57%
majority of whites say this, a
difference of 35 percentage
points.
The differences between
black officers and blacks
overall is significantly
smaller. About three-in-ten
black officers (29%) say the
necessary changes have been
made, a view shared by only
12% of blacks nationally and
a 17 percentage point
difference. By contrast, large
majorities of black officers and
blacks overall believe more
changes are needed (69% and
84%, respectively).

Police more likely than public to say that no more
changes are needed to give blacks equal rights with
whites
% saying that …
Officers

Public

Our country has made the
changes needed to give
blacks equal rights with
whites
Our country needs to
continue making changes to
give blacks equal rights with
whites

80
48
16
50

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016; survey of U.S.
adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

Police, public divided by race over whether attaining
equality requires more changes
% saying that …

Note: No answer category not shown.
Source: Survey of law enforcement officers conducted May 19-Aug. 14, 2016; survey of U.S.
adults conducted Aug. 16-Sept. 12, 2016.
“Behind the Badge”
PEW RESEARCH CENTER

www.pewresearch.org

89
PEW RESEARCH CENTER

Acknowledgments
This report is largely based on a Pew Research Center survey of police officers conducted by the
National Police Research Platform (NPRP), which collaborated with Center researchers on the
questionnaire and oversaw the technical administration of the survey. The Center retained final
editorial control over the questionnaire, the analysis of the survey and the contents of this report.
The NPRP, headquartered at the University of Illinois at Chicago from 2008 to 2016, seeks to
advance the science and practice of policing in the United States. It is now based at the Police
Foundation in Washington, D.C., with continued support from the National Institute of Justice. It
is managed by a team of leading police scholars and practitioners, supported by a respected
national advisory board. The NPRP researchers who participated in this project were:

Dennis P. Rosenbaum, University of Illinois at Chicago, NPRP Principal Investigator
Stephen D. Mastrofski, George Mason University, NPRP Co-principal Investigator
Susan M. Hartnett, University of Illinois at Chicago, NPRP Co-principal Investigator
Wesley G. Skogan, Northwestern University, NPRP Co-principal Investigator
Justin Escamilla, University of Illinois at Chicago, Research Coordinator
Georgina Enciso, University of Illinois at Chicago, Research Coordinator

This report is a collaborative effort based on the input and analysis of the following individuals at
the Pew Research Center. Find related reports online at pewresearch.org/socialtrends.

Kim Parker, Director of Social Trends Research
Rich Morin, Senior Editor
Renee Stepler, Research Analyst
Juliana Horowitz, Associate Director, Research
Claudia Deane, Vice President, Research
Courtney Kennedy, Director of Survey Research
Andrew Mercer, Senior Research Methodologist
Anna Brown, Research Analyst
Nikki Graf, Research Associate
Michael Keegan, Information Graphics Designer
Travis Mitchell, Digital Producer
Molly Rohal, Communications Manager
Brian Mahl, Communications Coordinator
Marcia Kramer, Kramer Editing Services

www.pewresearch.org

90
PEW RESEARCH CENTER

Methodology
Most of the data in this report come from online interviews completed by 7,917 law enforcement
officers from 54 police and sheriff’s departments across the United States. Of these, 6,795
interviews came from 43 municipal police departments and 1,122 interviews came from 11 county
sheriff’s departments. The surveys were administered between May 19 and Aug. 14, 2016. The
study was conducted by the National Police Research Platform (NPRP), a consortium of
researchers and practitioners around the country headquartered at the University of Illinois at
Chicago during the study period. The sample is designed and weighted to represent the population
of officers who work in agencies that employ at least 100 full-time sworn law enforcement officers
with general arrest powers. 15

Sample design
Participating agencies belong to a panel that was created by the NPRP. Because there is no
comprehensive national list of individual police officers, it is not possible to directly draw a simple
random sample of officers from all departments in the United States. Instead, selection needs to
take place in multiple stages. The first stage involves selecting a sample of police departments, for
which there is a comprehensive list. The second stage is to sample officers within those
departments, which is possible because departments have a clear accounting of the officers they
employ.
The first stage selection of police departments was performed by the NPRP in 2013. The 2007 Law
Enforcement Management and Administrative Statistics (LEMAS) survey was used as the
sampling frame. LEMAS is a survey of municipal police departments, sheriff’s departments and
state police. It is conducted periodically by the U.S. Bureau of Justice Statistics and includes all
agencies in the country with 100 or more sworn officers and a sample of agencies with fewer than
100 officers.
The NPRP panel was composed of a stratified random sample of agencies authorized to employ
100 to 3,000 officers and a purposive sample of larger agencies. For agencies other than the very
largest, the eligibility criteria for selection into the panel were different for municipal police
departments and sheriff’s departments. Municipal police departments were deemed eligible if they
were authorized to employ 100 or more full-time sworn officers and employed at least 50 officers
whose regular duties include responding to citizen calls for service. Sheriff’s departments were
eligible if they were authorized to employ between 100 and 3,000 full-time sworn officers whose
Because the number of officers employed in any given agency varies over time, the number employed at the time the NPRP panel was
created in 2013 may differ slightly from the number currently employed. One participating agency reported employing 97 full-time sworn
officers at the time the 2016 survey was conducted. All other participating agencies employed 100 or more.

15

www.pewresearch.org

91
PEW RESEARCH CENTER

duties do not include jail operations, court security (e.g., court bailiffs) or civil processes such as
serving subpoenas, and employed 50 or more officers who respond to citizen calls for service.
These criteria were intended to exclude sheriff’s departments that do not engage in traditional
policing duties. State police agencies were not eligible for inclusion.
Some 757 departments met these criteria. That list was stratified by agency type (police vs. sheriff),
number of law enforcement officers and Census Bureau region, and a random sample of 410
agencies was selected. Of these, 87 (21.2%) agreed to join the panel and participated in the first
NPRP survey. These 87 agencies (from the randomly selected sample) were supplemented with
four purposively selected larger municipal police departments that were authorized to employ over
3,000 full-time sworn officers. In 2013, these four agencies comprised roughly one-third of 11 total
comparably sized municipal police departments in the U.S. and employed 38.8% of the officers in
these departments. 16
Within participating departments, all eligible officers were invited to participate. Officers were
deemed ineligible only if their primary responsibilities consisted of court security, jail operations
or civil processes.

Data collection
In order to secure participation at the agency level for the current wave of the survey (2016), the
NPRP contacted the chief executives of each of the 91 previously empaneled agencies. Of these, 54
(59.3%) agreed to participate in the survey. Each agency that agreed to participate designated a
liaison to coordinate with the NPRP on the data collection effort. In consultation with NPRP staff,
agency liaisons and chiefs chose a starting date for the survey and developed department-specific
strategies, such as pre-notification emails, fliers or roll call announcements, for encouraging
officer participation.
Starting dates ranged from May 19 to July 11, 2016. On the starting date, department chiefs sent an
email to all eligible officers in their departments containing a link to the survey and inviting them
to participate. All eligible officers were assured that their participation was anonymous and that
their individual responses would not be accessible to anyone within their department. Agencies
were asked to have the chief send a follow-up email reminding officers to take the survey two
weeks after the start date, and a final reminder after three weeks. Data collection was closed after
approximately one month, with some agencies having shorter or longer field periods if requested
by the chief or liaison.
Source: LEMAS 2013. In 2013, the number of authorized full-time sworn officers was not measured. Comparable police departments are
defined as those employing over 2,600 full-time sworn officers.

16

www.pewresearch.org

92
PEW RESEARCH CENTER

Weighting
To ensure that estimates from this survey are generalizable to the national population of officers,
the data were weighted in a two-step process. First, each officer was weighted according to his or
her probability of selection. Because all officers within participating departments were sampled,
each officer’s probability of selection is equal to the probability that the individual’s department
was selected from within its stratum when the panel was initially created. The four large
departments that were selected purposively were treated as having been selected with certainty for
weighting.
Nonresponse to this survey could occur at three stages. First, agencies could have chosen not to
participate in the panel when they were first sampled. Second, some agencies that did join the
panel chose not to participate in this survey. Finally, within those participating agencies, not all
officers responded to the survey request. A total of 91 agencies belonging to the panel were invited
to participate in this survey. Of these, 54 (59.3%) agreed to participate. Across the 54 participating
agencies, the total number of eligible officers is 57,062, of whom 7,917 responded, for an officerlevel response rate of 13.9%. 17 The response rate varied by department size, at 32.7% for agencies
with fewer than 600 officers, 21.6% for agencies with 600 to 1,599 officers, and 7.9% for agencies
with 1600 or more officers. To adjust for nonresponse, an iterative technique that aligns the
sample to population benchmarks on a number of dimensions was applied. The data were
weighted by sex, race, rank, department type, department size in 2013, the size of the population
served in 2012 and U.S. Census Bureau region.
Parameters for these population characteristics come from the 2013 LEMAS survey of police
departments. Several questions that were used to define the group of eligible departments in 2007
were not asked in 2013. Equivalent parameters were calculated using an alternative definition for
the population of eligible departments that relied only on variables that were available in both
2007 and 2013. This revised population definition overlapped with the original definition for
97.9% of the officers in either set of departments.
Because LEMAS is a survey of departments rather than officers, officer-level estimates were
produced by weighting eligible departments by the number of officers who met the eligibility
criteria for this survey. For municipal police departments, this measure of size was equal to the
number of full-time sworn officers in each department. For sheriff’s departments, it was equal to
the number of full-time sworn officers whose primary responsibilities did not include jail
operations or court security. Because the 2013 LEMAS survey did not measure the number of
Because the number of eligible officers in each agency is available only for participating agencies, it is not possible to calculate a
cumulative response rate for all sampled agencies.

17

www.pewresearch.org

93
PEW RESEARCH CENTER

officers whose primary duty is civil processes, these officers could not be subtracted from the
measure of size. However, their share of the population in 2007 was too small to appreciably affect
weighting parameters (2.6% of officers in eligible sheriff’s departments and less than 1% of officers
in all eligible departments).

Law Enforcement Organizational (LEO) surveys
Several of the questions included in this survey were also asked on three earlier surveys conducted
by the National Police Research Platform referred to as LEO A, LEO B and LEO C, respectively.
The estimates in this report from these earlier LEO surveys were produced using data from only
those departments that participated in both the original LEO survey and this Pew Research Center
study. The original LEO survey datasets include officers who only partially completed the full
survey. To ensure comparability to the Pew Research Center survey, only officers who answered
one or more questions about age, sex, race or rank – which were located at the end of the survey –
were used for analysis. As with the current study, analysis was also restricted to officers whose
primary responsibilities did not include court security, jail operations or civil processes. The data
were weighted following the procedure described above.

Law Enforcement Organization Surveys
Survey
LEO A

Field Dates
July - Nov. 2013

Department Officer
response response
rate
rate
59.3%
16.8%

Departments
54

Officers
9,679

LEO B Sept. 2013-Jan. 2014

53

7,304

58.2%

12.7%

LEO C

48

7,115

52.7%

13.6%

Oct. 2014-Feb. 2015

Note: The number of departments and officers listed here do not reflect the full sample that
responded to each of the LEO surveys, but only those departments that participated in the
2016 Pew Research Center survey. The department-level response rates are based on the
91 departments that were invited to participate in the Pew Research Center study. The
officer response rate is based on the total number of eligible officers employed by the
departments who participated in both the LEO and Pew Research Center studies.
Source: National Police Research Platform.
“Behind the Badge”
PEW RESEARCH CENTER

Precision of estimates
Because of the complex design of this survey, it is not possible to produce a single margin of
sampling error that applies to all of the estimates from this survey. Some estimates (e.g., those
measuring the officer’s own attitudes) have a relatively small margin of error, while other

www.pewresearch.org

94
PEW RESEARCH CENTER

estimates (e.g., those measuring a department’s policies – which are roughly the same for
everyone in the department) have a larger margin of error. Consequently, we estimated precision
separately for each survey estimate. A list of the margins of sampling error for each question is
available upon request.
The multistage sampling process used in this survey – first departments and then officers – means
that survey responses are not independent of each other as they would be if we had taken a simple
random sample of all police officers directly. This clustering of officers within departments has the
effect of increasing the margin of error for survey estimates relative to a simple random sample of
the same size. Additionally, this clustering means that every question has a different margin of
error depending on how similar officers in the same department are to one another for the item in
question.
For questions where most officers in the same department give similar answers, the margin of
error is larger than for items where officers give more diverse responses. For example, question
39d asks officers if their department has modified its policies about the use of force. In principle,
the answer to this question should be the same for all officers in the same department. The margin
of sampling error for the percentage of officers who answered yes to this question is plus or minus
9 percentage points. For this item, each additional officer in a department contributes very little
additional information to the estimate. At the other end of the scale, the margin of sampling error
for the share of officers who see themselves as more of a protector than an enforcer in question 15
is plus or minus 1.5 percentage points. The sampling variability for other estimates falls
somewhere in between depending on how responses are distributed within departments.
In addition to clustering, the margin of error is affected by stratified sampling and weighting. The
analysis included in this report was performed using software that accounts for all of these
complex design features for tests of statistical significance and measures of sampling error.
One should also bear in mind that question wording and practical difficulties in conducting
surveys can introduce error or bias into survey results that is not captured by the margin of
sampling error.

Survey of the general public
The American Trends Panel (ATP), created by the Pew Research Center, is a nationally
representative panel of randomly selected U.S. adults living in households. Respondents who selfidentify as internet users and who provided an email address participate in the panel via monthly

www.pewresearch.org

95
PEW RESEARCH CENTER

self-administered Web surveys, and those who do not use the internet or decline to provide an
email address participate via the mail. The panel is being managed by Abt SRBI.
The data representing the general public in this report are drawn from the August wave of the
panel, conducted August 16-September 12, 2016 among 4,538 respondents (4,195 by Web and 343
by mail). The margin of sampling error for the full sample of 4,538 respondents is plus or minus
2.4 percentage points.
Members of the American Trends Panel were recruited from two large, national landline and
cellphone random digit dial (RDD) surveys conducted in English and Spanish. At the end of each
survey, respondents were invited to join the panel. The first group of panelists was recruited from
the 2014 Political Polarization and Typology Survey, conducted January 23rd to March 16th, 2014.
Of the 10,013 adults interviewed, 9,809 were invited to take part in the panel and a total of 5,338
agreed to participate. 18 The second group of panelists was recruited from the 2015 Survey on
Government, conducted August 27th to October 4th, 2015. Of the 6,004 adults interviewed, all
were invited to join the panel, and 2,976 agreed to participate. 19
Participating panelists provided either a mailing address or an email address to which a welcome
packet, a monetary incentive and future survey invitations could be sent. Panelists also receive a
small monetary incentive after participating in each wave of the survey.
The ATP data were weighted in a multi-step process that begins with a base weight incorporating
the respondents’ original survey selection probability and the fact that in 2014 some panelists were
subsampled for invitation to the panel. Next, an adjustment was made for the fact that the
propensity to join the panel and remain an active panelist varied across different groups in the
sample. The final step in the weighting uses an iterative technique that matches gender, age,
education, race, Hispanic origin and region to parameters from the U.S. Census Bureau's 2014
American Community Survey. Population density is weighted to match the 2010 U.S. Decennial
Census. Telephone service is weighted to estimates of telephone coverage for 2016 that were
projected from the July-December 2015 National Health Interview Survey. Volunteerism is
weighted to match the 2013 Current Population Survey Volunteer Supplement. It also adjusts for
party affiliation using an average of the three most recent Pew Research Center general public
telephone surveys. Internet access is adjusted using a measure from the 2015 Survey on
18 When

data collection for the 2014 Political Polarization and Typology Survey began, non-internet users were subsampled at a rate of 25%,
but a decision was made shortly thereafter to invite all non-internet users to join. In total, 83% of non-internet users were invited to join the
panel.
19 Respondents to the 2014 Political Polarization and Typology Survey who indicated that they are internet users but refused to provide an
email address were initially permitted to participate in the American Trends Panel by mail, but were no longer permitted to join the panel after
February 6, 2014. Internet users from the 2015 Survey on Government who refused to provide an email address were not permitted to join
the panel.

www.pewresearch.org

96
PEW RESEARCH CENTER

Government. Frequency of internet use is weighted to an estimate of daily internet use projected
to 2016 from the 2013 Current Population Survey Computer and Internet Use Supplement.
Sampling errors and statistical tests of significance take into account the effect of weighting.
Interviews are conducted in both English and Spanish, but the Hispanic sample in the American
Trends Panel is predominantly native born and English speaking.
The error attributable to sampling that would be expected at the 95% level of confidence for the
total sample is ± 2.4 percentage points.
Sample sizes and sampling errors for subgroups are available upon request.
In addition to sampling error, one should bear in mind that question wording and practical
difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.
The Web component of the August wave had a response rate of 81% (4,195 responses among 5,150
Web-based individuals in the panel); the mail component had a response rate of 76% (343
responses among 454 non-Web individuals in the panel). Taking account of the combined,
weighted response rate for the recruitment surveys (10.0%) and attrition from panel members who
were removed at their request or for inactivity, the cumulative response rate for the August ATP
wave is 2.9%. 20

Approximately once per year, panelists who have not participated in multiple consecutive waves are removed from the panel. These cases
are counted in the denominator of cumulative response rates.

20

www.pewresearch.org