The author(s) shown below used Federal funds provided by the U.S.
Department of Justice and prepared the following final report:

Document Title:

Final Technical Report Draft: Assessing Police
Use of Force Policy and Outcomes

Author:

William Terrill, Eugene A. Paoline III, Jason
Ingram

Document No.:

237794

Date Received:

February 2012

Award Number:

2005-IJ-CX-0055

This report has not been published by the U.S. Department of Justice.
To provide better customer service, NCJRS has made this Federallyfunded grant final report available electronically in addition to
traditional paper copies.

Opinions or points of view expressed are those
of the author(s) and do not necessarily reflect
the official position or policies of the U.S.
Department of Justice.

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

FINAL TECHNICAL REPORT DRAFT:
ASSESSING POLICE USE OF FORCE POLICY AND OUTCOMES

FUNDED BY THE NATIONAL INSTITUTE OF JUSTICE
GRANT NUMBER 2005-IJ-CX-0055

William Terrill
Principal Investigator
Michigan State University

Eugene A. Paoline III
Co-Principal Investigator
University of Central Florida

Jason Ingram
Project Site Director
Illinois State University

May 2011

This report was supported by Award No. 2005-IJ-CX-0055 by the National Institute of Justice,
Office of Justice Programs. U.S. Department of Justice. The opinions, findings, and conclusions
or recommendations expressed in this publication are those of the authors and do not necessarily
reflect the views of the Department of Justice.

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

ACKNOWLEDGMENTS
We owe an enormous amount of gratitude to the many individuals who contributed to our
research efforts. Without their support this project would never have been possible.
City of Albuquerque
Chief Ray Schultz, Sergeant Chad Dolan, Lieutenant Paul Feist, Lieutenant Allen Banks, Officer
Jack Coraci, Officer Duane Clark, Jay Davidson, Trey Flint, Erika Wilson, Anthony Simoes,
Veronica Pino, Valerie Jaramillo
City-County of Charlotte-Mecklenburg
Chief Darrel Stephens, Captain Charles Adkins
City of Colorado Springs
Chief Richard Myers, Lieutenant Arthur Sapp, Sergeant Mark Stevens, Officer Matthew Tyner,
Scott Haring
City of Columbus
Chief James Jackson, Commander Kent Shafer, Sergeant Jay Hammer, Officer Jon Gilles
City of Fort Wayne
Chief Russell York, Captain Lynn Wetmore, Captain Rod Howard, Lieutenant Lisa Williams,
Officer Juan Barrientes, Rick Wurm, Heidi McClain, Terry Wilson
City of Knoxville
Chief Sterling Owen, IV, Deputy Chief Don Green, Lieutenant Rick Ferguson, John Venn,
Diane Lewis
City of Portland
Chief Roseanne Sizer, Captain Eric Hendricks, Captain Randy Killinger, Sergeant Michael
Marshman, Bruce Cuthbertson, Christy Khalifa, Kristi Brant, Dorothy Brown, Merlyn Muse,
Leslie Stevens, Derek Reinke, Clair Warnock
City of St. Petersburg
Deputy Mayor Goliath Davis, Chief Charles Harmon, Major John Thompson, Major Phillip
Beahn, Lieutenant Gregory Totz, Rick Ferner, Beverly Woodall
American University
Dr. Edward Maguire

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Michigan State University
Dr. Edmund McGarrell, Dae-Hoon Kwak, Mike Rossler, Jason Hildebrant, Yi Ting Chua,
Marcia Williamson, Mike Tapling, Brian Buege, Kristin Bixman, Charles Scheer, Kathy
Harmon, Terri Bulock
Portland State University
Dr. Brian Renauer, Ryan Arnold, Sasha Craft, Shafina Fazal, Michael Hansen, Julie Nehl,
Alyssa Tibbs
University of Central Florida
Dr. Robert Langworthy, Jessica Williams, Jessica Maykrantz, Heidi Bonner, Sara Jane Shipman,
Amy Byrne, Aramis Donell, Scott Gardiner, Jayme Holmes, Ghazal Mehrani, Kendal Tidwell,
Jennifer Wing
University of North Caroline-Charlotte
Dr. Joseph Kuhns, Heather Bersot, Tammatha Clodfelter, Courtney Morrison, Diana Summers

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

TABLE OF CONTENTS

EXECUTIVE SUMMARY................................................................................................ i
CHAPTER I
Introduction.......................................................................................................................... 1
Literature Review............................................................................................................ 4
Gap in the Literature........................................................................................................ 8
CHAPTER 2
Project Overview.................................................................................................................. 10
Research Goals and Objectives........................................................................................ 10
Research Design............................................................................................................. 11
CHAPTER 3
Phase I................................................................................................................................... 12
Purpose............................................................................................................................. 12
Methodology.................................................................................................................... 12
Sampling Frame ....................................................................................................... 12
Sample Selection ...................................................................................................... 13
Survey Distribution and Response............................................................................ 14
Analyses and Findings..................................................................................................... 15
Continuum Design.................................................................................................... 15
Tactical Placement.................................................................................................... 17
Summary................................................................................................................... 32
CHAPTER 4
Phase II.................................................................................................................................
Purpose............................................................................................................................
Methodology...................................................................................................................
Agency Qualification and Selection..........................................................................
Data Collection..........................................................................................................
Agency Comparisons..................................................................................................
Policy Descriptions.....................................................................................................
Columbus...............................................................................................................
Charlotte-Mecklenburg..........................................................................................
Portland..................................................................................................................
Albuquerque...........................................................................................................
Colorado Springs...................................................................................................
St. Petersburg.........................................................................................................
Fort Wayne.............................................................................................................
Knoxville...............................................................................................................
Data Elements.............................................................................................................

35
35
35
36
38
40
42
42
43
44
45
46
47
48
49
56

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 5
Patrol Officer Survey Data...................................................................................................
Methodology...................................................................................................................
Survey Population......................................................................................................
Survey Protocol..........................................................................................................
Survey Response Rates..............................................................................................
Analyses and Findings....................................................................................................
Less Lethal Force Policy Perceptions................................................................
Summary................................................................................................................
Outcomes of Less Lethal Force Policy..............................................................
Summary................................................................................................................
Training and Less Lethal Force Policy..............................................................
Summary................................................................................................................
Background and Demographics.................................................................................
Summary................................................................................................................

58
58
58
59
62
64
64
68
69
73
74
79
80
90

CHAPTER 6
Force Report Data................................................................................................................
Methodology...................................................................................................................
Analyses and Findings....................................................................................................
Frequency of Force by Workload Measures..............................................................
Force and Resistance Types.......................................................................................
Bivariate Analyses.....................................................................................................
Summary................................................................................................................
Multivariate Analyses................................................................................................
Summary................................................................................................................
Suspect Injuries and Force.........................................................................................
Officer Injuries and Force..........................................................................................

91
91
94
94
96
107
127
127
134
135
139

CHAPTER 7
Citizen Complaint Data........................................................................................................ 145
Methodology.................................................................................................................... 145
Analyses and Findings..................................................................................................... 147
Frequency of Complaints............................................................................................ 147
Summary................................................................................................................. 149
Disposition of Complaints.......................................................................................... 151
Summary................................................................................................................ 154
Disciplinary Actions................................................................................................... 155
Summary................................................................................................................. 161
Frequency of Complaints by Workload Measures...................................................... 161
Summary................................................................................................................. 166
Use of Force and Discourtesy Officer Totals............................................................. 167
Summary................................................................................................................. 170
Multivariate Analyses................................................................................................. 170
Summary................................................................................................................ 171

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 8
Civil Litigation Data.............................................................................................................
Methodology...................................................................................................................
Analyses and Findings....................................................................................................
Summary.........................................................................................................................
CHAPTER 9
Project Summary Findings...................................................................................................
Phase I.............................................................................................................................
Force Continuum Policies.........................................................................................
Phase II............................................................................................................................
Patrol Officer Survey Data.........................................................................................
Less Lethal Force Policy Perceptions............................................................
Outcomes of Less Lethal Force Policy..........................................................
Training and Less Lethal Force Policy..........................................................
Summary................................................................................................................
Force Report Data......................................................................................................
Aggregate Officer Force Usage in Relation to Calls for Service, Reported
Crimes, and Arrests................................................................................................
Aggregate Officer Force.........................................................................................
Aggregate Suspect Resistance................................................................................
Force Relative to Resistance..................................................................................
Force and Resistance Comparisons........................................................................
Suspect Injuries and Force......................................................................................
Officer Injuries and Force.......................................................................................
Summary.................................................................................................................
Citizen Complaint Data..............................................................................................
Frequency of Complaints.......................................................................................
Disposition of Complaints.....................................................................................
Disciplinary Actions of Complaints.......................................................................
Complaints in Relation to Calls for Service, Reported Crimes, Arrests,
and Force................................................................................................................
Use of Force and Discourtesy Officer Totals.........................................................
Multivariate Analysis.............................................................................................
Summary................................................................................................................
Civil Litigation Data..................................................................................................
Number of Force Related Lawsuits Filed....................................................
Number of Force Related Lawsuits Settled or Lost.....................................
Monetary Payout for Force Related Lawsuits .............................................
Summary................................................................................................................

174
174
175
179

180
180
180
181
181
182
186
188
191
192
192
194
194
196
198
199
200
202
203
203
206
206
207
208
210
211
212
212
213
213
213

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 10
Discussion............................................................................................................................
Police Administrators......................................................................................................
Officer Concerns........................................................................................................
Citizen Concerns........................................................................................................
Overall Department Concerns....................................................................................
Police Researchers...........................................................................................................
Study Challenges........................................................................................................
Study Limitations.......................................................................................................
Future Research..........................................................................................................

214
214
216
217
218
219
219
221
224

REFERENCES................................................................................................................... 226
APPENDICES.................................................................................................................... 231

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

EXECUTIVE SUMMARY
While force continuum policies are frequently discussed in the policing literature by
academics and practitioners alike, little is known concerning how many agencies actually use a
continuum. Moreover, given potential variation in both continuum design (e.g., linear, matrix,
wheel, etc.) and tactical placement (e.g., OC spray, CED, etc.), even less is known with respect
to whether differences in continuum policies matter, and if so, in what way. Within this context,
this project set two separate, but interrelated goals: to identify the extent of variation in use of
force policies being used by police agencies throughout the country, and determine whether
certain types of policies offer more beneficial outcomes to police practitioners. Stated more
directly, the project sought to (1) identify existing variation in use of force policies, particularly
those employing a force continuum approach and (2) determine which types of policies offer
more beneficial outcomes to police practitioners, which are measured in terms of the degree to
which varying policies: (a) provide officers assistance and guidance with respect to force
decision-making, and (b) are associated with less force (i.e., by amount and type), injuries to
suspects and officers, citizen complaints, and lawsuits levied for improper force. To accomplish
these goals the project was divided into two primary phases.
PHASE 1
In the initial phase, we administered a mail survey to a stratified random sample of police
agencies across the country, based on agency size (i.e., sworn officers) and type (i.e., municipal,
sheriff). The purpose of the survey was to decipher the types of force policies that exist. The
key elements of the survey captured whether an agency had a written policy on non-lethal force,
employed a force continuum approach within their policy, the form or type of continuum used,

i

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the placement of various tactics within the continuum framework, and report and review
mechanisms. For greater detail see Chapter 3.
Key Findings
<

Over 80 percent of the responding agencies use some type of force continuum
policy:
<

<

<

With respect to the tactical placement of force tactics (soft hands, pain
compliance controls, hard hands) and weapons (batons, chemical sprays, CEDs),
and how police agencies rank the order of such in terms of progression, the key
finding uncovered was the amount of variation present:
<

In terms of force progression, a total of 123 different permutations were
uncovered, ranging from three to nine different levels (82.0% relied on
five or six levels).

<

In terms of citizen resistance progression, a total of 23 permutations were
uncovered, ranging from three to seven different levels (92.4%, relied on
five or six levels).

For departments that use a force continuum and provide officers with explicit
guidance as to the types of force most appropriate given varying types of
resistance (i.e., link citizen resistance to force), it was difficult to identify a
typically used preference:
<

<

of these, 73 percent rely on a linear design, followed by matrix/box
designs and circular/wheel designs at 10 percent each.

The most frequent approach was used by only 20 percent of the
departments, while the second and third most frequently used approaches
were just half of that at 10 percent.

The placement of chemical sprays and CEDs offered the greatest challenge for
police administrators as to the proper placement within the force continuum:
<

Roughly 30 percent of the agencies place chemical sprays with pain
compliance techniques, another 30 percent of the departments place
chemical sprays with hard hand tactics, and just over a third of the
agencies place chemical sprays with impact weapons.

<

Compared to chemical sprays, there was somewhat less variation when it
comes to CED placement, but far from a clear consensus. Nearly 60
ii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

percent of the agencies place CEDs at the impact weapon level, while
another 2.0 percent place it along with deadly force. The remaining
agencies place CEDs with some sort of hands on force (a quarter of the
agencies place CEDs at the same level as hard empty-hand tactics, with
another 13.1 percent placing CEDs with pain compliance techniques).
Summary
Overall, these findings are interesting given recent discussion within the literature
concerning the potential negatives of force continua in general, and linear designs in particular
(see Aveni, 2003, Peters and Brave, 2006, Petrowski, 2002, Williams, 2002). More specifically,
some have argued against the use of force continua on a number of fronts, such as in relation to
hampering decision making, fear of liability issues, and being more restrictive than the law.
Nonetheless, what is apparent from the findings presented here is that a large majority of police
agencies do incorporate a force continuum into their policy (80%), and the preferred model is
linear by design (73%).
Moreover, it was difficult to identify a standard practice that is used by police
departments across the country. While some departments are quite restrictive in terms of
allowing officers to use more severe forms of force only on actively aggressive suspects, other
agencies are quite liberal and place a large amount of discretion in officers hands by allowing
them to use nearly all types of force against nearly all types of resistance faced short of extreme
imbalance (e.g., allowing a baton strike to a compliant suspect). The most frequent approach is
used by only 20 percent of the departments, while the second and third most frequently used
approaches are just half of that at 10 percent. In essence, there really is no “commonly” used
means of tactical placement in terms of force continuum policies (i.e., where various forms of
hands on and weapons should be placed in relation to varying forms of suspect resistance).

iii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Departments pick and choose, and tweak and adapt, in a multitude of ways - all unfortunately,
with no empirical evidence as to which approach is best or even better than another.
PHASE 2
Based on results of the agency survey, eight agencies were selected for deeper
exploration as part of the second phase of the project (Columbus, OH, Charlotte-Mecklenburg,
NC, Portland, OR, Albuquerque, NM, Colorado Springs, CO, St. Petersburg, FL, Fort Wayne,
IN, and Knoxville, TN). Agency selection was based on ensuring adequate variation in terms of
differing policy types, but also a degree of comparability based on jurisdictional size, crime rate,
workload, and socioeconomic status. Upon securing agreements with police administrators, the
research team conducted multiple site visits over the course of two years at each agency. Data
collection consisted of a survey to patrol officers to assess their views on the impact of the
agency’s force policy on decision-making. In addition, use of force, citizen complaint, and civil
litigation data, as well as accompanying sources of information (e.g., organizational charts,
rosters, rules and regulation manuals, number of reported crimes, arrests, calls for service) were
collected retrospectively for 24 months. Further, a series of informal interviews were conducted
with officials at the middle and upper management levels. Collectively, these sources of data
served as the basis for the analyses and findings presented in this report. For greater detail see
Chapters 4 through 9.
Key Findings
First, it is important to note the extent of variation in the different policy approaches
(e.g., tactical placement of various forms of physical and weapon based force, and how citizen
resistance is, or is not, linked to force options in each of the eight policies). Two of the

iv

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

departments (i.e., Colombus and Fort Wayne) only offer force progression levels - with no
explicit link to citizen resistance. However, these agencies still vary in terms of force tactical
placement. For instance, Columbus calls for officers to use soft hands/pain compliance,
followed by chemical spray, CED, and then hard hands, all at separate and distinct levels. Fort
Wayne’s approach is slightly altered. First, it places chemical spray at the same level as soft
hands and pain compliance rather than on its own level. Second, the CED is placed at the impact
weapons level, which is after hard hands rather than before.
Six of the agencies, in one form or another, link force options to varying levels of citizen
resistance. Once again, the degree of variation is evident. Charlotte-Mecklenburg places
chemical sprays, hard hands, and the CED at different levels, while Albuquerque places all these
force types (as well as soft hand tactics and pain compliance) on the same force level. However,
it gets somewhat more complicated when citizen resistance is factored into the equation. In
Charlotte-Mecklenburg, chemical spray is at the defensively resistant level, while impact force
and the CED use requires greater that just defensive resistance suspects. By contrast,
Albuquerque permits soft and hard hand, pain compliance, chemical spray, and CED force on
verbally and defensively resistant suspects, reserving impact force to actively resistant suspects.
Other variations in the policies are also present. Portland places chemical sprays and the
CED together in the middle of the continuum against defensive citizen resistance, while St.
Petersburg places CEDs near the top of the continuum just prior to deadly force and requiring
active resistence. Knoxville instructs officers to use soft hands and pain compliance on verbal
and passive resisters, and then all other force short of deadly on defensive and active resisters.

v

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Perhaps one of the more interesting agency approaches is found in Colorado Springs.
One of the stated reasons for using a wheel, as opposed to a linear, type model is so officers do
not think about force progression in a ladder or step-by-step format. In fact, their wheel design
goes so far as to ensure that no force progression is evident (e.g., lethal force is placed next to
chemical agent on one side and soft control techniques on the other side). Nonetheless, in the
text of the policy citizen resistance is laid out in progression format and explicitly linked to
appropriate force options.
Below we offer a brief description of each agency model followed by key findings
presented according to how agencies compared to one another according to seven outcomes (see
also summary rankings in Chapter 9 for many additional findings).
1. The extent of force used in relation to workload factors (i.e., calls for service, arrests,
reported crime).
2. The extent to which officers use force proportional to the resistance they encounter:
3. The extent to which citizens are injured during forceful encounters.
4. The extent to which officers are injured during forceful encounters.1
5. The extent to which line level officers have a favorable view toward their agency’s
policy in terms of utility (i.e., assistance, clarity, guidance, restrictiveness, fairness).
6. The extent to which line level officers have a favorable view toward their agency’s
policy in terms of pre- and in-service force training.
7. The extent to which citizens file complaints for improper force/discourtesy in relation
to sworn personnel/workload.

1

Note that Albuquerque did not capture officer injuries as part of their reporting process.

vi

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Columbus
Columbus used a linear design referred to as the Action-Response to Resistance/
Aggression (Use of Force) model, which did not graphically depict a force continuum but rather
simply laid out eight levels of force (soft hands, chemical spray, electronic devices, hard hands,
impact weapon, canine, less lethal munitions, deadly) (see also Appendix B and agency policy
descriptions in Chapter 4).
More Positive Outcomes2
<
<
<
<
<

Columbus officers used the least amount of force relative to citizen resistance.
Columbus had the lowest amount of officer injuries.
Columbus had the second lowest amount of citizen injuries.
Columbus officers reported the second highest satisfaction rate in force training.
Columbus officers reported the fourth highest satisfaction rate in policy utility.

Less Positive Outcomes
<
<

Columbus had the second highest rate of citizen complaints for improper
force/discourtesy.
Columbus had the third highest rate of force in relation to workload.

Charlotte-Mecklenburg
Charlotte-Mecklenburg used a linear design, although their policy directive specifically
stated that it is “not designed to be a step by step progression.” The model graphically depicted
six levels of citizen resistance in linear fashion (cooperative, verbal and non verbal, passive,
defensive, active aggression, aggravated active) on a horizontal axis that sits-page43 above
another horizontal axis that depicts seven levels of force in linear fashion (professional presence
and verbal dialogue and commands, soft hands, chemical spray, hard hands, conducted energy

2

We identify more and less positive outcome categories based on whether the agency was in the upper half or lower
half ranking comparatively. More specifically, each agency was ranked one through eight: classifications for more
positive outcomes meant the agency scored in the top four, while classifications for less positive outcomes meant the
agency scored in the bottom four.

vii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

devices, impact weapon, and deadly force). These force options, while depicted on a horizontal
axis are also moving vertically as well. Further, there is not a complete overlap between
resistance and force options (e.g., hard hands can be used somewhere between defensive and
active resistance) (see also Appendix B and agency policy descriptions in Chapter 4).
More Positive Outcomes
<
<
<

Charlotte-Mecklenburg had the second lowest rate of force in relation to
workload.
Charlotte-Mecklenburg had the third lowest rate of citizen complaints for
improper force/discourtesy.
Charlotte-Mecklenburg officers reported the fourth highest satisfaction rate in
force training.

Less Positive Outcomes
<
<
<
<

Charlotte-Mecklenburg had the highest amount of citizen injuries.
Charlotte-Mecklenburg had the second highest amount of officer injuries.
Charlotte-Mecklenburg officers used the fourth highest amount of force relative
to citizen resistance.
Charlotte-Mecklenburg officers reported the fourth lowest satisfaction rate in
policy utility.

Portland
Portland employed a linear force continuum design that laid out and linked citizen
resistance with force options. This model depicted seven levels of citizen resistance (but not in
graphical format) in linear fashion (compliant, compliant/verbal, passive, physical, physical
aggressive, physical aggressive armed, deadly) in a right hand column, with seven force level
options in linear fashion in a middle column (officer presence, verbal, control holds, chemical
spray and Taser, baton and strikes/kicks, less lethal munitions, firearms), with five force levels
listed in the left hand column (presence, verbal, physical control, impact munitions, deadly) (see
also Appendix B and agency policy descriptions in Chapter 4).

viii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

More Positive Outcomes
<

Portland had the second lowest amount of officer injuries.

Less Positive Outcomes
<
<
<
<
<
<

Portland officers used the highest amount of force relative to citizen resistance.
Portland had the second highest rate of force in relation to workload.
Portland officers reported the second lowest satisfaction rate in policy utility.
Portland had the fourth highest amount of citizen injuries.
Portland officers reported the fourth lowest satisfaction rate in force training.
Portland had the fourth (tie) highest rate of citizen complaints for improper
force/discourtesy.

Albuquerque
Albuquerque employed a linear force continuum design (called the Reactive Control
Model or RCM) that laid out and linked citizen resistance with force options. This model
graphically depicted four levels of citizen resistance in linear fashion (cooperative, noncooperative, unarmed assailant, armed assailant) on a horizontal axis that sites above another
horizontal axis that depicts four levels of force in linear fashion (alert, control, active, survival).
Within each of the four broad levels there are specific types listed (e.g., under active it instructs
officers to consider anything from verbal commands to the use of an intermediate weapon) (see
also Appendix B and agency policy descriptions in Chapter 4).
More Positive Outcomes
<
<
<
<
<

Albuquerque officers reported the second highest satisfaction rate in policy
utility.
Albuquerque had the second lowest rate of citizen complaints for improper
force/discourtesy.
Albuquerque had the third lowest amount of citizen injuries.
Albuquerque officers reported the third highest satisfaction rate in force training.
Albuquerque had the fourth lowest rate of force in relation to workload.

ix

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Less Positive Outcomes
<

Albuquerque officers used the third highest amount of force relative to citizen
resistance.

Colorado Springs
Colorado Springs relied on a “Situational Force Model” (alternatively referred to as a
“wheel” model) as its means of a force continuum. This model graphically depicted an officer
standing in the middle of a circle with various force options surrounding him/her. The force
options are placed in random order to indicate there is no natural progression of force (e.g.,
deadly force is placed next to soft hand tactics). There is no graphical depiction of citizen
resistance as to which types of force are most appropriate given different types of resistance
(although the policy narrative/text does make a link to some degree, thereby indicating a degree
of linear progression despite the circular model approach) (see also Appendix B and agency
policy descriptions in Chapter 4).
More Positive Outcomes
<
<

Colorado Springs had the lowest rate of force in relation to workload.
Colorado Springs officers used the fourth lowest amount of force relative to
citizen resistance.

Less Positive Outcomes
<
<
<
<
<

Colorado Springs had the highest rate of citizen complaints for improper
force/discourtesy.
Colorado Springs officers reported the lowest (tie) satisfaction rate in force
training.
Colorado Springs had the third highest amount of citizen injuries.
Colorado Springs officers reported the third lowest satisfaction rate in policy
utility.
Colorado Springs had the third highest amount of officer injuries.

x

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

St. Petersburg
St. Petersburg relied on the Response to Resistance (Use of Force) matrix. This model
graphically depicted six levels of citizen resistance in linear fashion on a vertical axis (presence,
verbal, passive physical, active physical, aggressive physical, aggravated physical), with six
force level options in linear fashion on a horizontal axis (officer presence, communication,
physical control, intermediate weapons, incapacitating control, deadly force). The appropriate
force types are then linked to various resistance types by a check mark in boxes (e.g., if a citizen
displays aggressive physical resistance offices can use up to incapacitating control force (see
also Appendix B and agency policy descriptions in Chapter 4).
More Positive Outcomes
<
<
<
<

St. Petersburg had the lowest amount of citizen injuries.
St. Petersburg had the lowest rate of citizen complaints for improper
force/discourtesy.
St. Petersburg officers used the third lowest amount of force relative to citizen
resistance.
St. Petersburg had the third lowest amount of officer injuries.

Less Positive Outcomes
<
<
<

St. Petersburg officers reported the lowest satisfaction rate in policy utility.
St. Petersburg officers reported the lowest (tie) satisfaction rate in force training.
St. Petersburg had the fourth highest rate of force in relation to workload.

Fort Wayne
Fort Wayne employed a linear design, although the model did not graphically depict a
force continuum, but rather simply laid out six levels of force in linear fashion (officer presence,
verbal, soft hand, hard hand, intermediate weapon, deadly) and then partially identified types of
resistance via narrative/text (see also Appendix B and agency policy descriptions in Chapter 4).

xi

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

More Positive Outcomes
<
<
<

Fort Wayne officers reported the third highest satisfaction rate in policy utility.
Fort Wayne had the fourth lowest amount of citizen injuries.
Fort Wayne had the fourth lowest amount of officer injuries.

Less Positive Outcomes
<
<
<
<

Fort Wayne had the highest rate of force in relation to workload.
Fort Wayne officers used the second highest amount of force relative to citizen
resistance.
Fort Wayne officers reported the third lowest satisfaction rate in force training.
Fort Wayne had the fourth (tie) highest rate of citizen complaints for improper
force/discourtesy.

Knoxville
Knoxville’s policy graphically depicted five levels of citizen resistance in linear fashion
vertically (compliant, passive, active, aggressive, and assaultive/deadly) on the left side of the
continuum, with five force level options in linear fashion vertically on the right side of the
continuum (officer presence, empty hand soft/baton soft, compliance techniques, defensive
compliance techniques, deadly force), with five situational levels vertically in the middle of the
continuum (strategic, tactical, volatile, harmful, lethal) (see also Appendix B and agency policy
descriptions in Chapter 4).
More Positive Outcomes
<
<
<
<
<

Knoxville officers reported the highest satisfaction rate in policy utility.
Knoxville officers reported the highest satisfaction rate in force training.
Knoxville officers used the second least amount of force relative to citizen
resistance.
Knoxville had the third lowest rate of force in relation to workload.
Knoxville had the fourth lowest rate of citizen complaints for improper
force/discourtesy.

xii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Less Positive Outcomes
<
<

Knoxville had the highest amount of officer injuries.
Knoxville had the second highest amount of citizen injuries.

Top Rankings
For those interested in which agencies did the best and worst when considering the
varying outcomes, we list those agencies at the top and bottom of the rankings:
1. The extent of force used in relation to workload factors (i.e., calls for service, arrests,
reported crime):
Best Policies:
1. Colorado Springs
2. Charlotte-Mecklenburg
Worst Policies:
1. Fort Wayne
2. Portland
2. The extent to which officers use force proportional to the resistance they encounter:
Best Policies:
1. Columbus
2. Knoxville
Worst Policies:
1. Portland
2. Fort Wayne
3. The extent to which citizens are injured during forceful encounters:
Best Policies:
1. St. Petersburg
2. Columbus

xiii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Worst Policies:
1. Charlotte-Mecklenburg
2. Knoxville
4. The extent to which officers are injured during forceful encounters:
Best Policies:
1. Columbus
2. Portland
Worst Policies:
1. Knoxville
2. Charlotte-Mecklenburg
4a. The extent to which citizens and officers are injured, balancing citizen and officer
injuries in terms of lower and higher injuries):
Best Policies:
1. Columbus
2. St. Petersburg
Worst Policies:
1. Knoxville
2. Charlotte-Mecklenburg
4b. The extent to which officers use force proportional to the resistance they encounter,
while also considering the extent to which citizens and officers are injured:
Best Policies:
1. Columbus
2. Colorado Springs
Worst Policies:
1. Fort Wayne
2-tie. Portland
2-tie. Charlotte-Mecklenburg

xiv

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

5. The extent to which line level officers have a favorable view toward their agency’s policy in
terms of utility (i.e., assistance, clarity, guidance, restrictiveness, fairness):
Best Policies:
1. Knoxville
2. Albuquerque
Worst Policies:
1. St. Petersburg
2. Portland
6. The extent to which line level officers have a favorable view toward their agency’s policy in
terms of pre- and in-service force training:
Best Policies:
1. Knoxville
2. Columbus
Worst Policies:
1. Colorado Springs
2. St. Petersburg
7. The extent to which citizens file complaints for improper force/discourtesy in relation to
sworn personnel/workload:
Best Policies:
1. St. Petersburg
2. Albuquerque
Worst Policies:
1. Colorado Springs
2. Columbus

xv

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
As one reads through this executive summary (and the 200 plus pages of the final report)
undoubtedly, like ourselves, there may be a sense of letdown. Ideally, one would want to walk
away from this study knowing which use of force policy (or policies) is the best and which
policy (or policies) should be avoided. Unfortunately, one of the greatest strengths of the
empirical approach taken (i.e., examining multiple outcomes) is also potentially its greatest
weakness. As such, we cannot unanimously endorse or condemn one use of force policy over
another.
What is abundantly clear from the many analyses and rankings conducted is that there is
no ideal (or flawed) policy approach across all outcomes. The good news is that we provide
empirical evidence of various strengths and weaknesses across many important police outcomes.
We leave it to police executives to consider those outcomes most important or relevant to them
and their constituents, and see which policy approaches performed more favorably in those
respects. As just a single example, if one is looking to reduce citizen injuries and complaints as
top policing concerns, St. Petersburg’s approach may be worth considering. However, one has
to also be cognizant that officers generally did not view St. Petersburg’s policy in a favorable
light. A “cop’s cop” police leader may thus prioritize such officer-related concerns and dismiss
St. Petersburg’s policy, and instead endorse Knoxville’s approach that received outstanding
feedback from patrol officers (irrespective of the fact that Knoxville patrol officers were injured
at the highest rate of all cities). Any number of other examples could be given as well with other
cities. In the end, one must weigh the advantages and drawbacks of each policy approach
against various desirable (or undesirable) outcomes. Of course, readers must also use caution

xvi

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

interpreting the findings presented throughout this summary and the full report. Although this is
the most comprehensive less lethal use of force study conducted to date, as one astute reviewer
accurately points out - we have just begun to scratch the surface with respect to how varying
types of policies may influence varying types of outcomes.

xvii

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 1
Introduction

Society permits the police to use force in the course of their duties. At the same time,
limitations are set defining the extent of coercive power that police may exercise in maintaining
order and enforcing laws. Such restrictions have two interrelated components. The first stems
from the U.S. Supreme Court via Graham v. Connor, 490 U.S. 86 (1989), which states that force
at arrest must be “...objectively reasonable in view of all the facts and circumstances of each
particular case...” Despite direction from the high court, determining force that is “objectively
reasonable” is not an easy task (Terrill and Paoline, 2010). As a result, police departments rely
on a second component, a use of force policy, to establish parameters for the application of force
and to offer more explicit direction to officers about what may be considered appropriate force.3
Within this context, such policies are sometimes linked to force continuum (Terrill, 2001).4 A
standard force continuum policy ordinally ranks varying levels of force and resistance in terms
of severity, with the explicit purpose of offering officers guidance on how to respond to specific
forms of resistance. Thus, force continuum policies attempt to more readily clarify what may be
considered objectively reasonable force.
Over the years, several variations incorporating the notion of a force continuum have
been proposed. One of the earliest, the linear design, is modeled in the form of a ladder or
hierarchical steps. According to McEwen, this type of “continuum approach is to rely first on
3

The importance of written policy has long been recognized. The 1967 United States President’s Commission on
Law Enforcement and the Administration of Justice Task Force Report went so far as to state that “[p]olicies should
be formulated to bar not only unnecessary force but describe, to the extent possible, the amount of force which is
permissible for making arrests and carrying out other police activities” (p. 183).

4

McEwen notes that force policies incorporating a continuum approach “…are much more likely than other policies
to address the fundamental issue of physical force (open hand control, fists, use of body, etc.)” (1997: p. 50).

1

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the officer’s presence to quell a situation, and if that fails, to move to increasingly severe types
of force” (1997: p. 49). Another type of a continuum structure is often referred to as a modifiedlinear design, where subject resistance is placed into one of several levels and force options for
escalation (and de-escalation) are presented within each level (see Connor, 1991). Other designs
are laid out in matrix form or depicted by a wheel (see Hoffman et al., 2004). In the matrix
approach, varying forms of suspect resistance are presented along rows while varying police
responses are offered on a horizontal axis. The wheel design is often presented by a series of
concentric circles, with the situation as the core, followed by an intermediate circle of suspect
resistance behaviors, followed by an outer circle displaying varying appropriate forms of police
force given the type of resistance. Regardless of what continuum approach a particular agency
chooses to use (if any), the placement of different forms of force within the continuum structure
can vary.
While force continuum policies are frequently discussed in the policing literature by
academics and practitioners alike, little is known concerning how many agencies actually use a
continuum of some sort, and if so, what type (e.g., linear, wheel, etc.). Perhaps more
importantly, the extent to which such policies work or result in some tangible beneficial outcome
is unknown. Does a force continuum policy actually guide officers in their force usage? Does a
continuum policy enhance the level of control an officer has on a suspect? What impact does a
continuum policy have on the types of force officers use, injuries, citizen complaints, and
lawsuits for police misbehavior?
Complicating matters further, as illustrated above, is the fact that there is no ideal or
commonly accepted ranking of force, either by researchers or practitioners (US GAO, 2005). A

2

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

continuum policy in one department can be completely different than a policy in another
department. Some agencies employ a more linear type of force continuum, while others opt for a
wheel type design. Some departments have non-specific types of policies whereby force and
resistance are referred to only in vague terms and not categorized into specific levels, while
others offer very detailed policies that lay out many levels of both force and resistance.
Relatedly, there is variation as to where specific types of less-lethal technologies should be
placed on the continuum scale. For example, some departments place Conducted Energy
Devices (CED) such as the TASER® low on the continuum (e.g., right after verbal direction),
while other departments place it high on the continuum scale (e.g., just before deadly force) (US
GAO, 2005). Hence, some agencies view CED use as more of a “first resort,” while others
encourage CEDs use as a “last resort,” although there is no empirical evidence available
indicating which approach is used with greater frequency.
To muddy the waters even further, there is also some debate as to whether force
continuum policies actually help or hinder decision-making. There is clearly support for
continuum policies from practitioners (Rogers, 2001), but such support is not universally shared.
For instance, Petrowski (2002: p. 1) notes “[t]he force continuum purports to provide a
mechanical application when officers should be making a subjective threat assessment” (see also
Aveni, 2003, Peters and Brave, 2006, and Williams, 2002). Hence, while some police agencies
may incorporate a force continuum approach within their policy, there is some debate as to
whether such an approach actually serves to enhance officer decision-making.
Given the different policy approaches used by police agencies, combined with a lack of
knowledge concerning effectiveness, the present study investigates the extent of variation in use

3

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

of force policies being used by police agencies throughout the country, and whether certain types
of policies offer more beneficial outcomes to police practitioners. Such outcomes include the
extent to which different policies provide officers assistance and guidance with respect to force
decision-making, and the link between varying types of policy approaches and the amount/type
of force used, as well as injury rates (both suspect and officer), citizen complaints, and lawsuits
levied for improper force. In short, we seek to determine if certain types of policies are in effect
better, when considering a wide range of factors. Finally, we believe studies that examine any
particular tactic or less-lethal technology (e.g., TASER®) apart from the broader application of
such tactics within varying policy types unnecessarily limits potential insight. It is the merging
of varying policy types with varying tactics, which include less-lethal technologies, that offers
the most potential and ultimately useful information for policy makers and practitioners.

Literature Review
Although force continuum policies are specifically designed to provide officers guidance
toward controlling suspects with an appropriate degree of force, previous research has generally
neglected this area of inquiry, and no known study has specifically investigated whether certain
types of continuum policies are more effective than others.5 A review of the literature
demonstrates just how little is known with respect to the interplay between force continuum
policy approaches and tangible outcomes. The studies that have touched on the notion of force

5

McEwen (1997) offers the best look at use of force policy development in his review of 96 agencies through the
use of a convenience sample.

4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

continuum frameworks that are most pertinent to the present study can be broken down into two
broad areas and are briefly reviewed below.6
One area of research has focused on the impact of specific force tactics located on a force
continuum scale. These studies are generally most interested in the effect of certain tactics (i.e.,
incapacitative abilities and injuries rates). Chemical irritant sprays such as Oleoresin Capsicum
(OC) have perhaps received the greatest amount of inquiry (see Kaminski et al., 1999). In
general, there is a fair amount of agreement that OC spray offers a moderate to high degree of
incapacitative effect with fairly low concern for long term injuries; however, there is some
debate as to whether the use of OC spray should be used in response to verbal resistance (see
Adang and Mensink, 2004, as well as Smith and Alpert, 2000).
A more contemporary concern revolves around the use of another technology - CEDs
(e.g., TASER®). Evidence in relation to CEDs, in terms of effectiveness (i.e., high control, low
injury), is currently emerging. Some of the reports illustrate detrimental effects. For example,
Amnesty International (2004) reported that 74 people died in the previous three years after being
shocked by a TASER®. Findings from other studies have reported some beneficial outcomes
associated with CED use. For example, Smith and colleagues (2007) looked at CED use in two
agencies and its effect on suspect and officer injuries. In one of the two agencies the use of
CEDs reduced the number of officer and suspect injuries, as well as the severity of injuries for
suspects. In the second agency the use of CEDs had no effect on injuries, although interestingly
pepper spray did. On the whole, beyond research on the effects of OC spray and the more

6

A third area of research that incorporates the force continuum involves determining the frequency in which officer
use varying types of force (from low level to more severe forms) and citizen resistance behaviors, as well as
investigating the causes or predictors of police use of force. For example, see Crawford and Burns (1998), Klinger
(1995), and Terrill and Mastrofski (2002).

5

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

limited work on CEDs, evidence concerning the extent to which other types of tactics offer high
control and low injury, including physical forms of control (e.g., pain compliance techniques,
martial art forms) and intermediate or less-lethal weapons (e.g., baton, pepperball, bean bag) has
been lacking.7
A second area of research has focused more on the principles underlying force continuum
policies, rather than the effectiveness of certain types of tactics located within a continuum
structure (e.g., effectiveness of a chemical irritant spray). For instance, in 1997 Alpert and
Dunham proposed the use of “force factor” scores (subtracting the highest level of resistance
from the highest level of force within individual police-suspect encounters) to assess relative
degrees of force in relation to citizen resistance. Applying this approach with official data
gathered from the Miami-Dade, Florida and Eugene/Springfield, Oregon police departments, the
researchers found that officers in Miami-Dade used less force than the level of resistance faced,
compared to officers in Eugene/Springfield who used more force than the level of resistance
faced (on average). Of particular note, this type of pattern mirrored how officers were instructed
and trained to use force. Such a finding provides preliminary evidence that a department’s
policy environment may have an impact on police officer use of force behavior. In addition, this
work demonstrated how agencies may be compared in relation to force, albeit in a limited
fashion (i.e., only two departments).
Using observational data from the Project on Policing Neighborhoods (POPN) study in
Indianapolis and St. Petersburg, Terrill (2001, 2003, 2005), in a series of works, built upon the
idea of force factor scores by assessing not only the highest level of suspect resistance and police
use of force within an individual incident, but all instances of resistance and force that occur, as
7

For one limited review see Bertomen, 2003.

6

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

well as other factors present during police-suspect encounters, beyond suspect resistance, that
might account for force usage. In this respect, Terrill was able to move closer to determining if
force usage could be more appropriately labeled “objectively reasonable” given the totality of
circumstances and not just the resistance behavior presented. This early work was the first to
examine how police officers and citizens respond to one another, in a temporally sequenced
manner, within the context of a force continuum approach; and was the first attempt to quantify
and analyze the interactive nature of multiple police-citizen encounters in this type of microoriented process since Sykes and Brent’s (1983) pioneering work. Terrill found that officers did
not leap at the opportunity to apply force on a resistant citizen, but generally went to great
lengths to resolve such encounters in the most non-coercive manner possible (often giving
suspects multiple opportunities to comply). Conversely, the issue of appropriateness was more
problematic when officers dealt with non-resistant suspects, as they were more likely to use
higher forms of force than the resistance posed by suspects.
Fortunately, other researchers have since offered additional studies that utilize a
somewhat parallel approach. Alpert and Dunham (2004), once again using official data from the
Miami-Dade police department, followed Terrill’s (2001) call for further refinement regarding
an interactive approach and the application of force. Their work offered a second look at how
police officers and citizens respond to one another in terms of force and resistance as policecitizen encounters evolve. One of the findings, according to the authors, was that there was a
greater likelihood of police use of force the longer an encounter lasted, even when dealing with
non-violent citizen resistance. They also reported that “[p]reemptive force seemed to be
effective and produced an overall decrease in the number of incidents involving the use of force.

7

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Once a cycle of force is initiated, however, there appears to be only a limited opportunity to deescalate the level of force” (Alpert and Dunham, 2004: p. 122). Interesting, Terrill (2001) found
somewhat contradictory results with the POPN data, in that preemptive force led to greater uses
of force later in the encounter and that officers were actually quite skilled at de-escalation when
citizens raised the stakes via resistant behavior. Such findings only lead to an even greater need
for additional research.

Gap in the Literature
While researchers have looked at the effect of varying types of tactics located on a force
continuum scale, and the nature of police use of force behavior in terms of the underlying
principles of a continuum approach, prior work has not examined the different types of policies
currently being used by police agencies and the broader application of such policies.
Little is known concerning the extent to which police agencies across the country actually
incorporate a force continuum within their policy, and if so what type.8 Moreover, given
potential variation in both design (e.g., linear, matrix, wheel, etc.) and tactical placement (e.g.,
OC spray, CED, etc.), even less is known with respect to whether differences in continuum
policies matter, and if so, in what way. For instance, virtually nothing is known about the
connection between force continuum policies and tangible outcomes such as officers’
perceptions of their policy, the types of force used to control citizens, the nature of injuries to
officers and suspects, the rate of complaints generated, or the number of lawsuits filed for

8

According to Alpert and Dunham (2004) the Police Executive Research Forum (PERF) conducted a national
survey of police use of force in 1998. The authors report that 72 percent of responding police agencies rely on a
force continuum policy, but do not identify the types of force continuum policies being used either in terms of
frequency or variation.

8

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

allegations of improper force. In essence, previous research has failed to examine the
association between certain types of force continuum policies and beneficial (or detrimental)
outcomes. The present study begins to fill this gap.

9

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 2
Project Overview

Research Goals and Objectives
The project has two separate, but interrelated goals: to identify the extent of variation in
use of force policies being used by police agencies throughout the country, and determine
whether certain types of policies offer more beneficial outcomes to police practitioners. Stated
more directly, the project seeks to:
(1) identify existing variation in use of force policies, particularly those employing a
force continuum approach;
(2) determine which types of policies offer more beneficial outcomes to police
practitioners, which are measured in terms of the degree to which varying policies:
(a) provide officers assistance and guidance with respect to force decisionmaking, and
(b) are associated with less force (i.e., by amount and type), injuries to suspects
and officers, citizen complaints, and lawsuits levied for improper force.

10

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Research Design
The project was divided into two primary phases in connection with our two stated goals.
In the initial phase we administered a mail survey to a stratified random sample of 1,083 police
agencies across the country, based on agency size (i.e., sworn officers) and type (i.e., municipal,
sheriff). The purpose of the survey was to decipher the types of force policies that exist. More
directly, the key elements of the survey captured whether an agency employed a force continuum
approach within their policy, the form or type of continuum used, and the placement of various
tactics within the continuum framework.
Based on results of the agency survey, eight agencies were selected for deeper
exploration as part of the second phase of the project. Agency selection was based on ensuring
adequate variation in terms of differing policy types, but also a degree of comparability based on
jurisdictional size, crime rate, workload, and socioeconomic status. Upon securing agreements
with police administrators, the research team conducted multiple site visits over the course of
two years at each agency. Data collection consisted of a survey to patrol officers to assess their
views on the impact of the agency’s force policy on decision-making. In addition, use of force,
citizen complaint, and civil litigation data, as well as accompanying sources of information (e.g.,
organizational charts, rosters, manpower allocations, rules and regulation manuals, training
records, number of reported crimes, arrests, calls for service) were collected retrospectively for
24 months. Further, a series of informal interviews were conducted with officials at the middle
and upper management levels. Collectively, these data sources served as the basis for the
analyses and findings presented in this report.

11

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 3
Phase I
Purpose
Despite the attention force continuum policies have garnered within the policing
literature (Alpert and Dunham, 1997, 2004; Aveni, 2003; Connor, 1991; Crawford and Burns,
1998; Garner et al., 1995, 2002; Hoffman, et al., 2004; Klinger, 1995; McEwen, 1997;
Petrowski, 2002; Terrill, 2001, 2003, 2005; Williams, 2002), there has only been one known
attempt to determine how many agencies actually incorporate a continuum approach within their
policy (i.e., the 1998 PERF survey). More importantly, not a single study has attempted to
determine the varying types of force continuum policies being used in any systematic manner.
In fact, with respect to force policy in general, with the exception of reports indicating how many
agencies employ some sort of written policy on less lethal force, and varying forms of report and
review procedures (Alpert and Dunham, 204; Hickman and Reaves, 2003; Pate and Fridell,
1993), there is a substantial void within the empirical literature. Hence, the purpose of Phase I
was to design and administer a national survey to police agencies so as to assess the extent to
which different types of force polices exist, particularly those using a force continuum.

Methodology
Sampling Frame
As of 2000, there were 17,784 police agencies in the United States according to the
Census of State and Local Law Enforcement Agencies.9 With the assistance of Edward Maguire,

9

U.S. Dept. of Justice, Bureau of Justice Statistics. Census Of State And Local Law Enforcement Agencies
(CSLLEA), 2000: [U.S.] [Computer File]. Conducted by U.S. Dept. of Commerce, Bureau of the Census. 3rd ICPSR
ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2003.

12

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

project consultant, our sampling frame was comprised to include 7,306 agencies that (1) have
primary responsibility for policing a residential population, (2) employ 10 or more full-time
police officers, and (3) are a municipal or county agency (see Table 3-1). Selection based on
these three criteria include over 90 percent of all full-time sworn officers in the country
(Weisburd et al., 2000).
Table 3-1. Sampling Frame by Size and Type of Agency
Type

10-49

50-99

100-249

250-749

750+

Total

Police

4,104

745

378

125

57

5409

Sheriff

1,342

286

174

78

17

1897

Total

5,446

1,031

552

203

74

7306

Sample Selection
Because we lacked sufficient resources to survey the population of agencies meeting our
selection criteria, we needed to draw a representative sample of those agencies. Due to large
differences in the number of agencies in different size categories, we chose a disproportionate
stratified random sampling strategy based on agency size and type. For agency size, we chose
five categories based on number of full-time sworn officers with arrest powers. For agency type,
we established two categories. The first was police, formed by combining municipal, county,
and regional police into one category. The second was sheriff’s organizations. Once these
breakdowns were incorporated, we then selected 5 percent of agencies with 10-49 officers, 25
percent of agencies with 50-99 officers, 50 percent of agencies with 100-249 officers, and all
agencies in the next two size categories of 250-749 and 750 and more. This strategy resulted in
an overall sample size of 1,083 agencies. Table 3-2 shows the sample size within each stratum.

13

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-2. Sample Size by Size and Type of Agency
Type
Police
Sheriff
Total

10-49
205
67
272

50-99
186
72
258

100-249
189
87
276

250-749
125
78
203

750+
57
17
74

Total
762
321
1083

Survey Distribution and Response
Relying on well established techniques (see Dillman, 1978), the survey methodology
involved a multi-stage process beginning with pre-testing the instrument on 17 sworn officers
employed by 14 different police agencies located in five different states in January and February
of 2006. Upon receiving comments and making revisions, the final survey instrument was
mailed out to all 1,083 agencies in March of 2006.10 A second wave was distributed to nonresponding agencies in May of 2006, followed by a third wave in September of 2006. In an
attempt to further maximize the response rate, a phone call was made to every non-responding
agency between the second and third waves.11 Upon completion of wave three we had
accumulated a total of 591 responses for a response rate of 54.5 percent (wave 1=273, 25.2%;
wave 2=216, 19.9%; wave 3=102, 9.4%). In a final effort to ensure the greatest possible
response rate, a fourth wave was mailed out to the remaining 492 respondents in February 2007,
which resulted in another 71 responses. In total, we received 662 total responses (a response
rate of 61.1%).12
10

A personalized cover letter identifying the potential benefits of participation along with human subject protection
information accompanied each survey instrument.
11

In some instances we were asked to fax or mail another copy of the survey.

12

Nine additional surveys were received, but were filled out so minimally (e.g., answered only one or two questions)
or did not include a signature as required by our Institutional Review Boards (IRB), that they were not included in
the final count. In addition, 10 agencies contacted us (via a response letter or phone call/message) indicating that
they either did not wish to participate in our particular survey, or in some cases they do not participate in any survey

14

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

As illustrated in Table 3-3, in terms of agency size, within each of the five group
breakdowns the response rate progressively increases as the size of the agency increases. At the
ends of the spectrum the difference is most pronounced. While just under 40 percent of the
agencies with 10-49 sworn officers responded to the survey (105 responded of 272 surveyed),
over 90 percent of the largest (750 or more sworn officers) agencies responded (67 responded of
74 surveyed). In terms of agency type, police agencies were more likely to respond than sheriff
agencies. Of the 762 police departments surveyed, 494 responded (64.8%). For sheriff agencies,
168 of 321 responded (52.3%). Thus, from a generalizability standpoint, the survey results are
most reflective of mid-to-large sized police agencies.
Table 3-3. Responses by Type and Size of Agency
Type
Police
Sheriff
Total
Percent

10-49
82
23
105
38.6

50-99
109
30
139
53.9

100-249
145
40
185
67.0

250-749
107
59
166
81.7

750+
51
16
67
90.5

Total
494
168
662

Analyses and Findings
Continuum Design
The key aspect of the agency survey was to determine whether agencies use a force
continuum; and if so, the type of continuum design used and placement of tactics/weapons along
the continuum. Thus, we asked respondents if they relied upon a force continuum approach as
part of their less lethal force policy. We offered explicit direction so as to clarify our intent
stating, “[b]y force continuum, we mean a guideline (sometimes depicted graphically) that

research (it was noted that such decisions are sometimes arbitrary, while at other times the result of insufficient
personnel to devote appropriate time allocations).

15

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

officers can use to determine the type of force that may be used in generic situations. Such
guidelines are sometimes (but not always) linked with varying forms of citizen resistance in an
attempt to assist officers in matching the level of force to the level of resistance/threat
encountered. Some examples include linear (e.g., ladder, stair, FLETC), wheel, and matrix/box
designs, although there are many variations of continuum designs besides these few examples.”
Over 80 percent of the respondents indicated that they did rely on some form of a force
continuum.13 In comparison to the 1998 PERF survey, the percentage uncovered here is
somewhat higher. In that survey of 571 police agencies, according to Alpert and Dunham (2004:
p. 156), “...72 percent were based on a use-of-force continuum.”14
For those respondents that indicated the use of a continuum, we then asked them to
identify the “type” of force continuum their agency uses. We cautioned that there is no correct
or ideal continuum design in existence, but rather that some agencies simply prefer one design
over an another. To help guide them further we offered basic template examples (as illustrated
in Appendix A) of some of the continuum designs currently in existence. The results are shown
in Table 3-4. By far, the most prevalent response was the use of a linear design “without”
graphic representation (e.g., see Appendix A, question 2a for an illustration). The next most
frequent design was a linear approach “with” a graphic design (e.g., see Appendix A, question
2b for an illustration). Hence, nearly three-quarters of the respondent agencies (73.4%)
indicated the use of a linear design in some form. The next most frequently used designs
reported were a matrix/box approach (e.g., see Appendix A, question 2e for an illustration) and
13

Of the 662 responding agencies, 641 answered this question; of these, 518 indicated they used a force continuum
(80.8%).
14

McEwen (1997) found just 24 of 96 (or 25%) agencies in his review of force policies incorporated a force
continuum (he relied on a convenience sample).

16

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

some sort of circular/wheel approach (e.g., see Appendix A, question 2c and 2d for illustrations),
with both approaches used about 10 percent of the time (10.1% and 9.3% respectively). In sum,
our findings show that a substantial majority of police agencies do rely on a force continuum
structure, and that the most frequently used design type is linear in some fashion.
Table 3-4. Force Designs Frequency*
Continuum Design

N

%

Linear without Graphic Representation

240

46.5

Linear with Graphic Representation

139

26.9

Matrix/Box

52

10.1

Partial/Full Wheel

48

9.3

Other

37

7.2

*Of the 518 agencies that stated they use a force continuum, 516 indicated the type.

Tactical Placement
In addition to querying agencies about the type of force continuum used, we also asked
respondents to detail their force and citizen resistance progression by prompting them with
“[p]lease identify the progression of citizen resistance and police use of force outlined in your
continuum policy, from the lowest level to the highest level, and indicate the types/levels of
force recommended to officers for each type of resistance/threat they encounter (i.e., the range of
force options available to officers for each type of citizen resistance encountered)” (e.g., see
Appendix A, question 3 for an illustration). Of the 518 agencies that indicated they use a force
continuum as part of their policy, 476 of the respondents provided sufficient enough detail so as
to detail their force progression, while 371 agencies provided sufficient enough detail so as to
detail their resistance progression.
17

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

In an attempt to simplify progression patterns for illustrative purposes we placed the
numerous types of force and resistance into six categories. In relation to force, the categories
were broken into: officer presence/verbal direction, soft empty-hands (e.g., touching, pat down,
firm-grip, simple restraint), pain compliance techniques (e.g., pressure point controls), hard
empty-hands (e.g., hand strikes, punches, kicks, take downs without a weapon), impact weapons
(e.g., baton/ASP/flashlight strikes, pepperball, beanbag), and deadly force (e.g., handgun, rifle).
In addition to these six categories, chemical sprays (e.g., OC or pepper) and conducted energy
devices (CEDs) were also coded, but distinguished from the other six categories so as to present
the varying types of continuum placements different agencies use. In relation to resistance, the
categories were broken into: compliant, verbal (e.g., refusing verbal direction), passive (e.g.,
failing to respond to an officer/ignoring), physical defensive (e.g., bracing, pulling away,
fleeing), physical active (e.g., hostile and overt physical aggression towards the officer), and
deadly (e.g., attempt or actual attack that could cause death).
Even with an attempt to consolidate the many force and resistance types into broader
categories, a great deal of variation was uncovered in terms of how police agencies go about
detailing their policy in relation to the number of levels and placement of tactics. For instance,
of the 476 agencies that outlined their force progression, a total of 123 different permutations
were uncovered ranging from three to nine different levels (the vast majority, 390 or 82.0%,
relied on five or six levels).15 Of the 371 agencies that listed citizen resistance progression, a
total of 23 permutations were uncovered ranging from three to seven different levels (the vast
15

The number of permutations is reduced to 59 if one accounts for, or subsumes within, repeated forms of force
taking place at multiple levels. For instance, departments will often call for impact force at more than one level.
Thus, as an example, one department may place baton strikes at level five followed by an impact munition at level
six. If one counts these types of examples separately the permutation count rises to 123, if these are subsumed into
one level and labeled “impact force,” as is the case in many of the analyses presented here, the number is reduced to
59.

18

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

majority, 343 or 92.4%, relied on five or six levels).16 Within this context, we present the
findings from this section in a multitude of formats so as to depict the most comprehensive
picture of the varying ways to measure and understand how departments across the country
instruct officers via policy.
Excluding the 115 agencies that place chemical sprays (e.g., OC or pepper) and/or CEDs
(e.g., TASER®) on their own distinct level of force, Table 3-5 shows how the remaining 361
detail their force progression.17 First, looking at the ends of the continuum, a large majority of
the departments place both the least and most severe forms of force on their own level. Over 86
percent of the agencies place officer presence/verbal direction on a level by itself, while 97.8
percent place deadly force by itself. Nonetheless, some agencies do not place such rigid
restrictions on force placement at the ends of the continuum. For example, 49 of the agencies
place officer presence/verbal direction on the same level as soft hand tactics, while eight
departments permit a deadly force option along with non-lethal force options.

16

When repeated forms of resistance (i.e., meaning the same type of resistance types) are accounted for the number
of permutations is reduced from 23 to 15 (see Table 3-9).
17

Of the 59 permutations noted in footnote 14, there were 14 combinations uncovered when excluding agencies that
place chemical sprays and CEDs on their own distinct level. When these two additional force types are included, an
additional 47 permutations resulted across 115 agencies (most permutations obviously were only used by one or two
agencies). Aggregate results from the 115 agencies involving separate level chemical sprays and CEDs, and where
they fell on the continuum, are presented in Table 3-8.

19

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-5. Progression of Force
Factor

N

%

Level 1 - Presence/Verbal
Level 2 - Physical Soft
Level 3 - Pain Compliance, Physical Hard
Level 4 - Impact
Level 5 - Deadly

84

23.3

Level 1 - Presence/Verbal
Level 2 - Physical Soft
Level 3 - Pain Compliance
Level 4 - Physical Hard, Impact
Level 5 - Deadly

75

20.8

Level 1 - Presence/Verbal
Level 2 - Physical Soft, Pain Compliance
Level 3 - Physical Hard
Level 4 - Impact
Level 5 - Deadly

59

16.3

Level 1 - Presence/Verbal
Level 2 - Physical Soft, Pain Compliance, Physical Hard
Level 3 - Impact
Level 4 - Deadly

45

12.5

Level 1 - Presence/Verbal, Physical Soft
Level 2 - Pain Compliance, Physical Hard
Level 3 - Impact
Level 4 - Deadly

24

6.6

Level 1 - Presence/Verbal
Level 2 - Physical Soft, Pain Compliance
Level 3 - Physical Hard, Impact
Level 4 - Deadly

21

5.8

Level 1 - Presence/Verbal, Physical Soft
Level 2 - Pain Compliance
Level 3 - Physical Hard
Level 4 - Impact
Level 5 - Deadly

13

3.6

Level 1 - Presence/Verbal
Level 2 - Physical Soft
Level 3 - Pain Compliance, Physical Hard, Impact
Level 4 - Deadly

12

3.3

20

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Factor

N

%

Level 1 - Presence/Verbal, Physical Soft
Level 2 - Pain Compliance
Level 3 - Physical Hard, Impact
Level 4 - Deadly

11

3.0

Level 1 - Presence/Verbal
Level 2 - Physical Soft
Level 3 - Pain Compliance
Level 4 - Physical Hard
Level 5 - Impact
Level 6 - Deadly

7

1.9

Level 1 - Presence/Verbal
Level 2 - Physical Soft, Pain Compliance, Physical Hard
Level 3 - Impact, Deadly

7

1.9

Level 1 - Presence/Verbal
Level 2 - Physical Soft
Level 3 - Pain Compliance, Physical Hard, Impact, Deadly

1

0.3

Level 1 - Presence/Verbal
Level 2 - Physical Soft, Pain Compliance, Physical Hard, Impact
Level 3 - Deadly

1

0.3

Level 1 - Presence/Verbal, Physical Soft
Level 2 - Pain Compliance, Physical Hard, Impact
Level 3 - Deadly

1

0.3

Total

361

100.0

Tables 3-6 and 3-7 illustrate where chemical sprays and CED devices are placed when
such force types are found at the same level as one of the six other categories. Chemical sprays
and CEDs are analyzed in this manner because how or where they fit into force continuum
progression has not been entirely clear from past research. The findings presented here show
precisely where on the force continuum such mechanisms fall using a representative nationwide
sample. As shown in Table 3-6, placement of chemical sprays varies widely. Roughly 30
percent of the agencies place chemical sprays with pain compliance techniques, another

21

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

approximate 30 percent of the departments place chemical sprays with hard hand tactics, and just
over a third of the agencies place chemical sprays with impact weapons. A handful of
departments even place chemical sprays as low as empty-hand soft techniques (n=12) and as
high as deadly force (n=6). Such widespread variation indicates a tremendous lack of agreement
by police practitioners as to where chemical sprays should be placed on the force continuum.
Table 3-6. Force Level of Chemical Spray when Shared with Other Force
Factor

N

%

Physical Soft

12

3.3

Pain Compliance

110

30.5

Physical Hard

105

29.2

Impact

127

35.3

Deadly

6

1.7

Total

360

100.0

Similar to the approach used for chemical sprays, Table 3-7 depicts where CEDs are
placed when embedded with other types of force. Compared to chemical sprays there is
somewhat less variation, but far from a clear consensus. Nearly 60 percent of the agencies place
CEDs at the impact weapon level, with another two percent placing CEDs along with deadly
force. Thus, just under two-thirds of the departments require some type of hands on force before
resorting to the CED. However, over a third of the agencies place CEDs with some sort of hands
on force. In particular, a quarter of the agencies place CEDs at the same level as hard emptyhand tactics, with another 13.1 percent placing CEDs with pain compliance techniques.

22

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-7. Force Level of CED When Shared with Other Force
Factor

N

%

Physical Soft

2

0.8

Pain Compliance

32

13.1

Physical Hard

60

24.6

Impact

145

59.4

Deadly

5

2.0

Total

244

100.0

As noted previously, 115 of the 476 agencies that offered sufficient detail to determine
force progression place chemical sprays and/or CEDs on their own distinct force level. Table 38 shows where these weapons are placed on the continuum. The location of chemical sprays is
much more varied when compared with CEDs. The most frequent location for chemical sprays
is right after pain compliance techniques (36.8%), followed by hard empty-hand hard tactics
(29.4%). Just under 20 percent of the agencies place it prior to any form on hands on force (right
after officer presence/verbal force). With respect to CEDs, most agencies place this weapon near
the top of the continuum. Only one agency places it right after verbal force (and before handson), while 10 of the 13 agencies that place CED on its own level place it after hard hands or
impact. Finally, and somewhat interesting after the CED findings, is the placement of chemical
sprays and CEDs when they are both placed on the same level together, but by themselves on
their own level. In this case, the modal placement location is after pain compliance, but before
physical hard hands (42.1%). Further, over a quarter of the agencies use this placement prior to
pain compliance, and zero agencies after impact.

23

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-8. Force Level of Chemical Sprays and CED When Occupying Own Force Level*

Factor

Chemical Spray
N%

CED
N%

Both
N%

After Presence/Verbal

13

19.1

1

7.7

7

18.4

After Physical Soft

8

11.8

-

-

3

7.9

After Pain Compliance

25

36.8

2

15.4

16

42.1

After Physical Hard

20

29.4

5

38.5

12

31.6

After Impact

2

2.9

5

38.5

-

-

Total

68

100.0

13

100.0

38

100.0

* 7 agencies placed CED on it’s own level directly after Chemical Sprays (11 agencies had both chemical
spray and CED on their own, but separate levels).

Next, we turned to looking at how citizen resistance progression was laid out by agencies
with respect to the force continuum (see Table 3-9). Recall that 371 agencies listed their
resistance progression in sufficient detail so as to code this information. As shown in Table 3-9,
nearly half (n=179, 48.2%) of all the responding agencies who detailed their citizen resistance
progression use a five level layout in the following order: 1-compliant, 2-verbal/passive
combined, 3-physical defensive, 4-physical active, and 5-deadly. The second most frequently
used approach (n=123, 33.2%) is similar except that verbal and passive resistance are split and
placed on separate levels according to the following: 1-compliant, 2-verbal, 3-passive combined,
4-physical defensive, 5-physical active, and 6-deadly. These two approaches combined to
account for 302 (81.4%) of the 371 agencies. Thus, while there are 12 additional permutations
used, the drop off in frequency is dramatic (e.g., the third most frequently used progression
format only contains 16 departments).

24

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-9. Progression of Resistance
Factor

N

%

Level 1 - Compliant
Level 2 - Verbal, Passive
Level 3 - Physical Defensive
Level 4 - Physical Active
Level 5 - Deadly

179

48.2

Level 1 - Compliant
Level 2 - Verbal
Level 3 - Passive
Level 4 - Physical Defensive
Level 5 - Physical Active
Level 6 - Deadly

123

33.2

Level 1 - Compliant, Verbal
Level 2 - Passive
Level 3 - Physical Defensive
Level 4 - Physical Active
Level 5 - Deadly

16

4.3

Level 1 - Compliant, Verbal
Level 2 - Passive, Physical Defensive
Level 3 - Physical Active
Level 4 - Deadly

12

3.2

Level 1 - Compliant
Level 2 - Verbal, Passive, Physical Defensive
Level 3 - Physical Active, Deadly

9

2.4

Level 1 - Compliant
Level 2 - Verbal, Passive, Physical Defensive
Level 3 - Physical Active
Level 4 - Deadly

7

1.9

Level 1 - Compliant, Verbal, Passive
Level 2 - Physical Defensive
Level 3 - Physical Active
Level 4 - Deadly

6

1.6

Level 1 - Compliant
Level 2 - Verbal
Level 3 - Passive, Physical Defensive
Level 4 - Physical Active
Level 5 - Deadly

6

1.6

25

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Factor

N

%

Level 1 - Compliant
Level 2 - Verbal, Passive
Level 3 - Physical Defensive, Physical Active
Level 4 - Deadly

4

1.1

Level 1 - Compliant
Level 2 - Verbal
Level 3 - Passive
Level 4 - Physical Defensive
Level 5 - Physical Active, Deadly

2

0.5

Level 1 - Compliant
Level 2 - Verbal
Level 3 - Passive
Level 4 - Physical Defensive, Physical Active
Level 5 - Deadly

2

0.5

Level 1 - Compliant
Level 2 - Passive
Level 3 - Verbal
Level 4 - Physical Defensive
Level 5 - Physical Active
Level 6 - Deadly

2

0.5

Level 1 - Compliant
Level 2 - Verbal, Passive, Physical Defensive, Physical Active
Level 3 - Deadly

1

0.3

Level 1 - Compliant, Verbal, Passive
Level 2 - Physical Defensive, Physical Active
Level 3 - Deadly

1

0.3

Level 1 - Compliant, Verbal, Passive, Physical Defensive
Level 2 - Physical Active
Level 3 - Deadly

1

0.3

Total

371

100.0

We then examined the extent to which police agencies connect varying levels and types
of citizen resistance to varying levels and types of police force. As presented in Table 3-10, of
the 476 agencies indicating that they incorporate a force continuum approach into their policy,

26

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

125 (26.3%) do not instruct officers as to the types (or range) of force to be used in relation to
the resistance faced.18
Table 3-10. Agencies that Link Force to Resistance Via a Continuum
Does Agency Link Force to Resistance?

N

%

Not Linked

125

26.3

Semi-Linked

15

3.2

Linked

336

70.6

Total

476

100.0

At the other end of the spectrum are those agencies that use both a resistance and force
continua (336, 70.6%). These agencies link specific types of force to specific types of
resistance.19 All apply some sort of “out-clause” in the sense that there is no requirement that
officers “progress” up or down force continuum levels in strict form (e.g., that officers must
exhaust all lower forms of force prior to moving up the continuum), but all do specify the range
of force that should be used given the level of resistance posed by the suspect.20
The top 10 most frequently used approaches are presented in Table 3-11. The top
approach is used by just 20 percent of the departments, and the second and third most frequently
used approaches are half of that at just 10 percent. The fourth approach drops in half again
18

Of these 125, 105 just specify force progression within their policy while 20 specify both resistance and force but
make no connection, or such a vague connection, indicating that force is not linked to resistance levels.
19

Somewhat interesting, 21 of the 48 (or 43.8%) departments that use a partial or full wheel/circular continuum
design indicated a specific connection or link between force and resistance. The presumed reason for adopting a
wheel type approach is often because the agency does not want to “lock-in” officers as to what force to use given a
level of resistance. While such a goal may be accomplished given the graphic depiction of a partial or full wheel
model, in these 21 agencies the policy still makes the connection anyway via the text of the policy.
20

As indicated in Table 3-10, 15 (3.2%) departments were coded as semi-linking force to resistance. These agencies
provide some partial guidance as to the force-resistance relationship, but are loosely coupled (for example, a graphic
illustration in the policy may depict impact weapons most closely connected to active resistance, but also partially
connected to defensive and passive resistance).

27

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

(n=17, 5.0%). The remaining approaches all have less than 10 agencies using them (over half of
all agencies). This speaks to the enormous amount of variation in existence when it comes to
force policy. Many different agencies are using many different approaches. In essence, there
really is no “commonly” used means of tactical placement in terms of force continuum policies
(i.e., where various forms of hands on and weapons should be placed in relation to varying forms
of suspect resistance). Departments pick and choose, and tweak and adapt, in a multitude of
ways - all unfortunately, with no empirical evidence as to which approach is best or even better
than another.

28

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 3-11. 10 Most Frequently Used Force Continuum’s
Level

Resistance

Force

N

%*

1
2
3
4
5

Compliant
Verbal, Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Physical Soft
Pain Compliance
Physical Hard, Impact
Deadly

68

20.2

1
2
3
4
5

Compliant
Verbal, Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Physical Soft
Pain Compliance, Physical Hard
Impact
Deadly

35

10.4

1
2
3
4
5
6

Compliant
Verbal
Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Presence/Verbal
Physical Soft, Pain Compliance
Physical Hard
Impact
Deadly

34

10.1

1
2
3
4
5
6

Compliant
Verbal
Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Physical Soft
Pain Compliance, Physical Hard Impact
Impact
Impact
Deadly

17

5.0

29

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Level

Resistance

Force

N

%*

1
2
3
4
5

Compliant
Verbal, Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Presence/Verbal
Physical Soft, Pain Compliance, Physical Hard
Impact
Deadly

9

2.7

1
2
3
4
5

Compliant
Verbal, Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal
Physical Soft, Pain Compliance
Physical Hard Impact
Impact
Deadly

9

2.7

1
2
3
4
5

Compliant, Verbal
Passive, Physical Defensive
Passive, Physical Defensive
Physical Active
Deadly

Presence/Verbal, Physical Soft
Pain Compliance, Physical Hard Impact
Impact
Impact
Deadly

9

2.7

1
2
3
4
5

Compliant, Verbal
Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal, Physical Soft
Pain Compliance
Physical Hard Impact
Impact
Deadly

6

1.8

1
2
3
4
5

Compliant
Verbal, Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal, Physical Soft
Pain Compliance
Physical Hard Impact
Impact
Deadly

5

1.5

30

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Level

Resistance

Force

N

%*

1
2
3
4
5
6

Compliant
Verbal
Passive
Physical Defensive
Physical Active
Deadly

Presence/Verbal, Physical Soft
Presence/Verbal, Physical Soft
Pain Compliance, Physical Hard Impact
Impact
Impact
Deadly

5

1.5

* Percent is calculated based on total number of agencies that link force to resistance in a force continuum format (N=336)

31

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The agency survey was designed to offer an in-depth look at the extent to which police
agencies incorporate a force continuum into their policy. While there has been limited
information about the extent to which police agencies rely on a force continuum within their
policy approach, until now we knew nothing about the extent of variation in terms of design type
and the tactical placement of verbal and physical force, as well as weapons. What follows is a
summary of the key findings from Phase I of the project.
We found that over 80 percent of agencies reported using some type of continuum. Of
these agencies, the linear design was, by far, the most frequently used (73%), followed by
matrix/box designs and circular/wheel designs, each with roughly 10 percent of the agencies
using them. This finding is somewhat interesting given recent discussion within the literature
concerning the potential negatives of force continua in general, and linear designs in particular
(see Aveni, 2003, Peters and Brave, 2006, Petrowski, 2002, Williams, 2002). More specifically,
some have argued against the use of force continua on a number of fronts, such as in relation to
hampering decision making, fear of liability issues, and being more restrictive than the law.21
Even during our sites visits as part of Phase II of this study we heard concerns from police
officials about the potential down side of using a force continuum. Nonetheless, what is
apparent from the findings presented here is that a large majority of police agencies do
incorporate a force continuum into their policy, and the preferred model is linear by design.
With respect to the tactical placement of force tactics (soft hands, pain compliance
controls, hard hands) and weapons (batons, chemical sprays, CEDs), and how police agencies
21

We even received written comments on some return surveys indicating that the force continuum was not part of
policy for fear of liability concerns, but it was used in training - and apparently in the views of administrators
somehow outside the scope of plaintiffs’ attorneys.

32

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

rank the order of such in terms of progression, the key finding uncovered was the enormous
variation present. Even with an attempt to consolidate the many force and resistance types into
broader categories, a great deal of variation was uncovered in terms of how police agencies go
about detailing their policy in relation to the number of levels and placement of tactics.22 Of the
476 agencies that outlined their force progression, a total of 123 different permutations were
uncovered ranging from three to nine different levels (the vast majority, 390 or 82.0%, relied on
five or six levels). Of the 371 agencies that listed citizen resistance progression, a total of 23
permutations were uncovered ranging from three to seven different levels (the vast majority, 343
or 92.4%, relied on five or six levels).
When looking at departments that attempt to incorporate a force continuum approach into
their policy, and provide officers with explicit guidance as to the types of force most appropriate
given varying types of resistance (i.e., link citizen resistance to force), it is difficult to identify a
typically used preference. The most frequent approach is used by only 20 percent of the
departments, while the second and third most frequently used approaches are just half of that at
10 percent. This speaks to the enormous amount of variation in existence when it comes to force
policy. Many different agencies are using many different approaches.
The placement of chemical sprays and CEDs appear to offer the greatest challenge for
police administrators as to the proper placement within the force continuum. Roughly 30
percent of the agencies place chemical sprays with pain compliance techniques, another 30
percent of the departments place chemical sprays with hard hand tactics, and just over a third of

22

Even in states with state level guidelines regarding policy on police use of force (e.g., Florida, Michigan, and New
Jersey) variation was present (i.e., individual agencies would take the state guidelines and tweak them to
accommodate their individual agency preference - such as moving the TASER® from one level to another).

33

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the agencies place chemical sprays with impact weapons. A handful of departments even place
chemical sprays as low as empty-hand soft techniques (n=12) and as high as deadly force (n=6).
Compared to chemical sprays there is somewhat less variation when it comes to CED placement,
but far from a clear consensus. Nearly 60 percent of the agencies place CEDs at the impact
weapon level, while another two percent place it along with deadly force. Thus, just under twothirds of the departments require some type of hands on force before resorting to the CED.
However, over a third of the agencies place CEDs with some sort of hands on force. In
particular, a quarter of the agencies place CEDs at the same level as hard empty-hand tactics,
with another 13.1 percent placing CEDs with pain compliance techniques.
In sum, while some departments were quite restrictive in terms of allowing officers to use
more severe forms of force only on actively aggressive suspects, other agencies were quite
liberal and placed a large amount of discretion in officers hands by allowing them to use nearly
all types of force against nearly all types of resistance faced short of extreme imbalance (e.g.,
allowing a baton strike to a compliant suspect). Overall, it is nearly impossible to identify a
standard approach being used by police agencies. In essence, there really is no “commonly”
used means of tactical placement in terms of force continuum policies (i.e., where various forms
of hands on and weapons should be placed in relation to varying forms of suspect resistance).
Departments pick and choose, and tweak and adapt, in a multitude of ways - all unfortunately,
with no empirical evidence as to which approach is best or even better than another.

34

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 4
Phase II

Purpose
While prior research has investigated many aspects of police use of force, little effort has
focused on attempting to link policy with outcomes. Thus, the second phase of the project was
designed to examine whether certain types of policies are more advantageous than others; this is
particularly relevant given the many different versions of force policy being used by police
agencies across the country, as illustrated in Phase I. More specifically, we seek to determine
which types of policies offer more beneficial outcomes to police practitioners, which are
measured in terms of the degree to which varying policies (a) provide officers assistance and
guidance with respect to force decision-making, and (b) are associated with less force (i.e., by
amount and type), injuries to suspects and officers, citizen complaints, and lawsuits levied for
improper force. It is our hope that results from this phase of the project will offer insight on
whether varying types of policies actually influence police use of force behavior and provide
practitioners with a guide for policy development by permitting informed decision-making based
on independent empirical assessment.

Methodology
While it would be ideal to examine force usage in each of the agencies surveyed on force
policies, practical limitations prohibited such an approach. As a result, and based on results from
the agency survey, eight agencies were selected for deeper exploration as part of the second
phase of the project.

35

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agency Qualification and Selection
Agency qualification was based on several criteria. First, agencies must have engaged in
the regular reporting of force via officer use of force reports.23 This data collection strategy
offers the most promising means of collecting large amounts of data in the most efficient
manner. It is also important to note that a growing number of police agencies presently mandate
use of force reporting (see Terrill et al., 2003).24 Second, agencies must have had a consistent
use of force policy and reporting procedure for two consecutive years, so as to ensure that
instrumentation is not a concern. The inclusion of agencies that have altered their force policy or
reporting procedures would not permit valid comparisons across agencies when attempting to
determine the identified outcomes. Third, we sought agencies that could reasonably be classified
as mid-to-large in size. We did not want agencies that were so small that obtaining a sufficient
number of force incidents would become an issue; and we did not want overly large agencies
(e.g., New York, Chicago, Los Angeles, Houston, Philadelphia) whereby generalizability would
become an issue. If these three thresholds were met then the crux of the agency selection
process involved building in variation in terms of force continuum design and tactical placement
on one hand, but also ensuring some degree of agency comparability based on jurisdictional size,
crime rate, and socioeconomic status.

23

Some agencies do no explicitly use the term “force reports,” but rather some variation thereof such as control of
persons/suspects reports, response to suspect resistance/aggression reports, supervisor control of persons/force/
resistance reports, among others. For our purposes we sought agencies that documented force incidents in some type
of special or separate report outside (or beyond) regular incident or arrest reports. We use the term force report to
generically speak for all such types of reports regardless of what unique name an agency might place on them.
24

The United States President’s Commission on Law Enforcement and the Administration of Justice dating back to
1967 noted the importance of systematic force reporting, “[s]uch policies can best be enforced if all officers who use
physical force for any reason are required to report in writing the circumstances under which the force was used” (p.
183).

36

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

From a “design” standpoint, our original plan was to select eight total agencies broken
down by two that use a linear approach, two that employ a wheel or circular type approach, two
with a matrix design, and two that use no force continuum within their policy framework.
However, as illustrated from the agency survey results, while over 80 percent of the agencies
utilize some form of a force continuum within their policy, 73.4 percent of them employ some
type of linear approach, with roughly 10 percent using both a wheel and matrix approach. As a
result, we wanted to over-sample (select) agencies using a linear approach, which more
adequately reflects departmental policies throughout the country rather than imposing an
artificial standard. Hence, we selected six departments using a linear style, one using a matrix
approach one using a wheel design. The only challenge we encountered was the inability to
identify and select at least one agency that incorporated no force continuum in their policy.
Though 19 percent of the total responding agencies stated they use no continuum, this was
almost universally reserved for the smallest agencies (i.e., those offering an inappropriate match)
and/or those that do not systematically capture force usage (i.e., use of force reports).25
Besides design considerations (e.g., linear, wheel, matrix), we wanted to ensure the eight
agencies varied in terms of “tactical placement.” Once again, as illustrated from the agency
survey results, even though many departments may use a linear type of design, the variation in
how this is depicted, and perhaps even more importantly, the progression of force as to how
officers are instructed, varies greatly. As such, we wanted to build in this variation.
Finally, we sought jurisdictions that were reasonably comparable from a socioeconomic
perspective. More directly, we paid close attention to attempting to select agencies that did not
25

There were two departments of appropriate size that use no continuum and collect force data, but when contacted
and requests were made for participation they declined.

37

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

differ substantially across measures such as race, unemployment, poverty, or crime rates. In
short, agency selection was based on ensuring adequate variation in terms of differing policy
types, but also a degree of comparability based on jurisdictional size, crime rate, and
socioeconomic status.
Once all the above considerations were made and tabulations were complete, 16 agencies
were identified and approached to determine possible participation. After numerous discussions
with administrators in these agencies eight participants were selected: Columbus, OH, CharlotteMecklenburg, NC, Portland, OR, Albuquerque, NM, Colorado Springs, CO., St. Petersburg, FL,
Fort Wayne, IN, and Knoxville, TN.

Data Collection
Upon securing agreements with police administrators, the research team conducted
multiple site visits over the course of two years at each agency. At a minimum, we spent at least
one full month on site at each location. On average, we traveled to each site four times and spent
a week per trip.26 Data collection consisted of a survey to patrol officers to assess their views on
the impact of the agency’s force policy on decision-making (see description below under officer
survey methodology subsection for further detail). In addition, use of force data, complaint data,
and civil litigation data, as well as accompanying sources of information (e.g., organizational
charts, rosters, rules and regulation manuals, number of reported crimes, arrests, calls for

26

Two of the sites (Fort Wayne and St. Petersburg) were within short driving distance from our home universities,
which allowed us to go back and forth more readily and therefore not always requiring full week trips at a time, (the
tradeoff was that we needed to make more than four trips).

38

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

service) were collected retrospectively for 24 months (hence, a two-year study period).27
Further, a series of formal and informal interviews were conducted with officials at the middle
and upper management levels. Collectively, these sources of data serve as the basis for the
analyses and findings presented in this report.
Overall, the data collection process was extremely labor intensive. First, with eight
police agencies involved we had to deal with eight different processes across a number of data
elements. For instance, for force data alone we had to uncover how and when officers file force
reports, how policy and/or training changed or was altered during our study time frame, how
force data are catalogued within the department, and how we would be able to obtain the data.
Similar types of information had to be gleaned and collected for a host of other data sources as
well, such as complaints, lawsuits, reported crimes, arrests, calls for service, staffing, along with
a number of other elements. In addition, since the project spanned the course of two years there
was some degree of turnover in terms of police personnel. In some cases our primary contact
liaison person within the police department was reassigned during the project, prompting us to
have to start the process anew with someone else that may have been unfamiliar with the project
at the time. In one city there was a change at the police chief level; fortunately, the new chief
allowed us to continue the project and provided resources to do so. In short, in no way was this a
project of minimal effort nor one that asked police departments to simply “send us data.” This is
critically important because ensuring we are comparing “common measures” in relation to force
usage, complaint generating, and the nature of civil litigation, cannot be done by simply asking

27

The exact two-year time frame for each site was as follows: Columbus (2006-2007 calendar years), CharlotteMecklenburg (2006-2007 calendar years), Portland (November 5, 2005-November 4, 2007), Albuquerque (April 13,
2006-April 12, 2008), Colorado Springs (2006-2007 calendar years), St. Petersburg (April 1, 2006-March 31, 2008),
Knoxville (June 1, 2005-May 31, 2007), and Fort Wayne (December 18, 2004-December 17, 2006).

39

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

police agencies to send us data. The onus must be on the researcher, not the police department,
to have any semblance and confidence that the data being used in the analyses that follow are as
clean and comparable as reasonably possible.

Agency Comparisons
Table 4-1 lists the eight study sites and how they compare across a multitude of factors
ranging from design type, to population, to crime rates, to various city characteristics. The size
of the cities in terms of population and number of sworn officers ranged from 733,203 and 1,819
in Columbus to 182,337 and 382 in Knoxville, respectively. Several of the cities are particularly
similar across these two dimensions (e.g., Columbus and Charlotte-Mecklenburg, Portland and
Albuquerque, Colorado Springs and St. Petersburg, Fort Wayne and Knoxville). In addition,
many of the socioeconomic indicators are relatively closely situated. For instance, percent nonwhite ranges from 32 percent in Columbus to just under 20 percent in Colorado Springs; poverty
rates range from six percent in Colorado Springs to 14 percent Knoxville. From a comparative
perspective, Rochester, NY, one of the 16 cities originally considered for study, has a non-white
percentage over 50 and a poverty rate of nearly 30; hence, it was not selected as a study city. We
tend to see a little more variation on some other measures. For example, while five of the eight
cities have crime rates per 1,000 population in the 60s and 70s, the high end of the spectrum is
Knoxville just under 82 percent, while Colorado Springs and Fort Wayne are in the mid-to-high
40s.

40

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 4-1. Study Sites
Columbus

Continuum Design

Charlotte Portland Albuquerque Colorado
Mecklenburg
Springs

St.
Petersburg

Fort
Wayne

Knoxville

Linear

Linear

Linear

Linear

Wheel

Matrix/Box

Linear

Linear

733,203

733,291

538,133

513,124

374,112

248,069

248,423

182,337

% Non-White

32.0

36.0

22.1

28.4

19.3

28.6

24.5

20.3

% Female Headed

9.3

7.6

6.3

8.0

7.1

7.9

9.8

8.0

% Below Poverty

10.8

6.6

8.5

10.0

6.1

9.2

9.6

14.4

% Unemployed

3.5

3.7

4.5

3.8

3.1

3.2

4.3

3.9

Part I Crimes/1,000 pop.

78.8

79.8

65.5

66.9

49.5

76.6

43.6

81.8

Total # Sworn Officers

1,819

1,638

989

986

669

520

457

382

# Officers/1,000 pop.

2.48

2.23

1.84

1.92

1.79

2.10

1.84

2.0

City Characteristics
Population

Agency Characteristics

41

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Policy Descriptions
As noted above, we have six agencies that use a linear design, one that uses a wheel
design, and one that uses a matrix approach. The following offers a description of each agency
policy. For additional detail, in terms of how agencies depict the force continuum within their
policy, as well tactical placement of force options, see also Table 4-2 and Appendix B.
Columbus28
Columbus used a linear design referred to as the Action-Response to Resistance/
Aggression (Use of Force) model, which did not graphically depict a force continuum but rather
simply laid out eight levels of force (soft hands, chemical spray, electronic devices, hard hands,
impact weapon, canine, less lethal munitions, deadly) . The policy does not identify or link
resistance types to specific types of force. The Action Response Report form is filled out every
time reportable force is used. There are separate addendums for mace, CEDs, and firearms. For
levels 0-1, the officer completes a use of force form and forwards it to their immediate
supervisor. If the use force level is Level 3 (CED) through 7 (less lethal control), or if someone
is injured (at any level of force), the supervisor conducts an investigation, reviews the use of
force report (initially started by officer), and recommends discipline (if any). EMS is called
every time a CED is used, as they are required to remove the probes. Level 8 – deadly force, the

28

The information used for each of the city/department sections was gathered from a variety of sources over a twoyear period and included formal and informal interviews/meetings with city and agency officials, and a vast array of
written materials from many different sources (e.g., written polices and procedures, agency memos, city websites,
etc.). Some of the written description is taken directly from prior and/or existing documents, while other material is
taken and pieced together from a variety of sources. As a result, we do not always apply traditional quotations
around every direct source and/or quote (admittedly some of the material is that of the city, agency, website, or other
source). Our primary task was to first understand the many processes occurring in each of the eight different cities
before being able to document such processes with the intent to provide a proper context to the analyses in the
report. Our goal was not to create and write a detailed historical and complete contemporary description focusing on
source specificity, but to offer a background setting. Within this context, we apologize to any author or entity whose
material may be presented here without any or full attribution.

42

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

officer essentially hands all of the investigating and report writing over to the immediate
supervisor.
Once completed, use of force reports are reviewed by the chain of command, while the
Deputy Chief makes the final determination regarding the findings. As mentioned, the front-line
supervisor plays a major role in the use of force investigation from beginning to end. Force
reporting also includes off-duty incidents involving sworn personnel. Internal Affairs keeps all
records of use of force (above level 1), while all use of force data is stored at the Training
Academy.
Charlotte-Mecklenburg
Charlotte-Mecklenburg used a linear design, although their policy directive specifically
stated that it is “not designed to be a step by step progression.” The model graphically depicted
six levels of citizen resistance in linear fashion (cooperative, verbal and non verbal, passive,
defensive, active aggression, aggravated active) on a horizontal axis that sits above another
horizontal axis that depicts seven levels of force in linear fashion (professional presence and
verbal dialogue and commands, soft hands, chemical spray, hard hands, conducted energy
devices, impact weapon, and deadly force). These force options, while depicted on a horizontal
axis are also moving vertically as well. Further, there is not a complete overlap between
resistance and force options (e.g., hard hands can be used somewhere between defensive and
active resistance). A Use of Force Report form is filled out every time reportable force is used.
When officers use a documented level of force they are required to contact their supervisor
immediately, who conducts an investigation of the incident (Supervisor Investigative Report).
The supervisor completes the report based on information provided by the officer(s) and

43

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

citizen(s) at the scene (via an investigation). If a CED is employed, a separate “Taser® Data
Download” report is completed. There are also addendums for OC spray and firearms.
Following the initial investigation of the use of force incident by the officer’s sergeant,
the chain of command (Division Captain, Service Area Major, and Internal Affairs) reviews the
incident, deeming it justified or unjustified use of force. The final disposition comes from the
Service Area Major, while Internal Affairs serves primarily as a check over the process for
consistency and oversight. Electronic tracking is done for each use of force incident, which
documents who reviewed the case, their comments, and for how long. There is a 45 day limit on
investigations of use of force incidents.
Portland
Portland employed a linear force continuum design that laid out and linked citizen
resistance with force options. This model depicted seven levels of citizen resistance (but not in
graphical format) in linear fashion (compliant, compliant/verbal, passive, physical, physical
aggressive, physical aggressive armed, deadly) in a right hand column, with seven force level
options in linear fashion in a middle column (officer presence, verbal, control holds, chemical
spray and TASER®, baton and strikes/kicks, less lethal munitions, firearms), with five force
levels listed in the left hand column (presence, verbal, physical control, impact munitions,
deadly). Every officer who uses reportable force must complete a Use of Force Report (UFR).
Primary officers involved in the incident are required to fill out all mandatory reports.
Narratives of the incident are typically not captured on the primary officer’s UFR because they
are written in the custody/incident reports. Thus, narrative data are not recorded electronically in
the Bureau’s data system (PPDS). The Bureau’s UFR was adopted in July 2004. Prior to this

44

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

time, the CED and less lethal force were captured on separate Bureau forms. Data on Firearm
Discharge incidents are captured on a separate form. Supervisors are not required to go to the
scene of every use of force incident; supervisors are directed to respond to the scene in instances
where a subject has been hit by a less lethal impact weapon or in instances where an officer has
negligently or unintentionally discharged the CED. Officers are required to notify supervisors
immediately whenever the baton, chemical spray, CED, or less lethal weapons are deployed on a
subject. In these cases, supervisors are supposed to document any injuries in an inter-office
memorandum, most often an After-Action Report, which are narratives that describe and
evaluate a police action. These reports, which are ultimately stored in the Chief’s Office, must
be submitted within a week to a Branch Chief who forwards them to Training and Internal
Affairs. Although supervisors are not required to go to the scene in all instances, they are
responsible for reviewing every UFR written by their officers.
Albuquerque
Albuquerque employed a linear force continuum design (called the Reactive Control
Model or RCM) that laid out and linked citizen resistance with force options. This model
graphically depicted four levels of citizen resistance in linear fashion (cooperative, noncooperative, unarmed assailant, armed assailant) on a horizontal axis that sites above another
horizontal axis that depicts four levels of force in linear fashion (alert, control, active, survival).
Within each of the four broad levels there are specific types listed (e.g., under active it instructs
officers to consider anything from verbal commands to the use of an intermediate weapon).
Every officer who uses reportable force must complete a Use of Force Report (UFR)
Bubblesheet (e.g., scantron like form) in addition to an Incident Report (Form 42). It is the

45

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

responsibility of the officer’s supervisor to ensure that the UFR Bubblesheets are filed when
reportable force is used, but they are not required to go to the scene. If an officer does not fill
out the report, the supervisor is responsible for filling it out. It is also the responsibility of the
supervisor to review all subordinate officers UFR Bubblesheet forms. The supervisor must sign
the form and note whether the force used was appropriate or if an investigation is warranted.29
Upon review of the supervisor, a copy of the UFR Bubblesheet and the Incident Report is
submitted for review to the Operations Review Lieutenant within 72 hours of the incident. The
Operations Review Lieutenant then distributes the reports to Internal Affairs, Risk Management,
the Legal Department, Tactical Teams, and the Academy when appropriate. The Operations
Review Lieutenant ensures that the report is reviewed by Legal and forwarded to Internal
Affairs. Finally, it is the Commander’s responsibility to ensure that all supervisors and
subordinates conform to the policy and that all reports are submitted to the appropriate units.
Colorado Springs
Colorado Springs relied on a “Situational Force Model” (alternatively referred to as a
“wheel” model) as its means of a force continuum. This model graphically depicted an officer
standing in the middle of a circle with various force options surrounding him/her. The force
options are placed in random order to indicate there is no natural progression of force (e.g.,
deadly force is placed next to soft hand tactics). There is no graphical depiction of citizen
resistance as to which types of force are most appropriate given different types of resistance
(although the policy narrative/text does make a link to some degree, thereby indicating a degree
of linear progression despite the circular model approach). Every officer who uses reportable
29

According to several department and city officials, prior to mid-2005, officers’ use of force reporting was
somewhat sporadic due to a broader organizational perception that filling them out was not important and that
supervisors had never needed to ensure that they were filled out.

46

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

force must complete a force report (called a Response To Aggression report or RTA). As part of
the policy, supervisors are required to respond to the scene; however, pending extreme cases
involving injury or some other circumstance a supervisor believes warrants, the supervisor does
not interview officers, witnesses, or the suspect. Rather, they are just present to oversee.
Once completed, the RTA is forwarded to an Administrative Sergeant who is the first to
review it, copy it, and provide a copy for the officer’s file. The RTA is then sent for review
through the chain of command (Shift Lieutenant, Division Commander, Office of Professional
Standards/Inspection Unit). RTA forms are stored with the Office of Professional Standards/
Inspection Unit.
St. Petersburg
St. Petersburg relied on the Response to Resistance (Use of Force) matrix. This model
graphically depicted six levels of citizen resistance in linear fashion on a vertical axis (presence,
verbal, passive physical, active physical, aggressive physical, aggravated physical), with six
force level options in linear fashion on a horizontal axis (officer presence, communication,
physical control, intermediate weapons, incapacitating control, deadly force). The appropriate
force types are then linked to various resistance types by a check mark in boxes (e.g., if a citizen
displays aggressive physical resistance offices can use up to incapacitating control force). The
CED has its own policy. Officers are required to fill out a Response to Resistance Report form
whenever any reportable force is used. In addition, officers are required to report when they
point their firearm at a suspect. The ASP, chemical sprays, and physical force are reported on
the Response to Resistance Form, but CEDs have their own form (which includes pointing and
accidental discharges. When CEDs are discharged, a supervisor comes to the scene where they

47

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

make sure the officer collects all supplemental information (i.e., CED form, download data, and
take pictures) is included. The supervisor is then required to review and sign off on the form onscene.
Once the Response to Resistance and/or the CED form is complete it, along with the
arrest report, is sent through the chain of command (Sergeant, Lieutenant, Major, Assistant
Chief) for review before it is forwarded to the Internal Affairs Division for final review. Internal
Affairs maintains a copy of force reports, providing a copy to the Training Division and the
original copy to the Records and Identification Division.
Fort Wayne
Fort Wayne employed a linear design, although the model did not graphically depict a
force continuum, but rather simply laid out six levels of force in linear fashion (officer presence,
verbal, soft hand, hard hand, intermediate weapon, deadly) and then partially identified types of
resistance via narrative/text. According to policy, force should be “reasonably necessary” and
used only after “discussion, negotiation, and persuasion have been found to be inappropriate or
ineffective.” Every officer who uses reportable force is required to fill out a use of force report
form in addition to an incident report. CED incidents and Firearm Discharges are reported on a
separate form, and kept with the force report/incident report forms.
After officers fill out use of force report forms, a sergeant reviews and signs off on them.
Supervisors are not required to go to the scene when force is used. The use of force report form
is then attached to the incident report and sent to the Office of Professional Standards, who then
sends a copy of the use of force and incident reports to the Training Center. A training
instructor, at the Training Center, reviews the use of force report to ensure that they are being

48

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

filled out by officers appropriately, and also to determine if the force utilized was effective or
not.
Knoxville
Knoxville’s policy graphically depicted five levels of citizen resistance in linear fashion
vertically (compliant, passive, active, aggressive, and assaultive/deadly) on the left side of the
continuum, with five force level options in linear fashion vertically on the right side of the
continuum (officer presence, empty hand soft/baton soft, compliance techniques, defensive
compliance techniques, deadly force), with five situational levels vertically in the middle of the
continuum (strategic, tactical, volatile, harmful, lethal). Officers are required to fill out a Use of
Force Report form at level two (empty hand soft, baton soft, contact controls, direction controls)
and above or if a suspect has (or complains) of an injury. There are no supplemental forms for
OC spray or the CED (but there are for K-9 and Firearm Discharge). Once force forms are filled
out, they are given to Sergeants for their review.30 CED usage is handled slightly differently, as
officers are required to notify a supervisor immediately31 and summon EMS medical assistance
for probe removal if they are embedded in the suspect’s skin. Supervisors are required to go the
scenes in which deadly force is used by an officer.
Once a use of force report is filled out (by the end of the shift) it goes through the chain
of command. Signatures are required from the immediate supervisor (who reviews the audio and
video tapes), a lieutenant, and a captain. A formal review of the use of force report is also
conducted by Internal Affairs. Reports are also reviewed by the Deputy Chief of Patrol, the
30

Informally, the department likes to have the supervisor present (not on scene) when the officer is filling out the
report to reduce errors in reporting.
31

Informally, the department encourages supervisors to go to the scene in CED cases, although it is not explicitly in
the policy.

49

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Chief, the Training Division, and CALEA (when they visit KPD), although signatures are not
required. Use of force reports are stored within the Internal Affairs Unit.
Table 4-2 offers a detailed layout of tactical placement and how citizen resistance is (or is
not) linked to force options for each of the eight policies (see also Appendix B for how the
policy is depicted graphically in each agency). As shown, each agency has a different approach
in some manner. Two of the departments (i.e., Colombus and Fort Wayne) only offer force
progression levels - with no explicit link to citizen resistance. However, these agencies still vary
in terms of force tactical placement. For instance, Columbus calls for officers to use soft
hands/pain compliance, followed by chemical spray, CED, and then hard hands, all at separate
and distinct levels. Fort Wayne’s approach is slightly altered. First, it places chemical spray at
the same level as soft hands and pain compliance rather than on its own level. Second, the CED
is placed at the impact weapons level, which is after hard hands rather than before.
Six of the agencies, in one form or another, link force options to varying levels of citizen
resistance. Once again, the degree of variation is evident. Charlotte-Mecklenburg places
chemical sprays, hard hands, and the CED at different levels, while Albuquerque places all these
force types (as well as soft hand tactics and pain compliance) on the same force level. However,
it gets somewhat more complicated when citizen resistance is factored into the equation. In
Charlotte-Mecklenburg, chemical spray is at the defensively resistant level, while impact force
and the CED use requires greater that just defensive resistance suspects. By contrast,
Albuquerque permits soft and hard hand, pain compliance, chemical spray, and CED force on
verbally and defensively resistant suspects, reserving impact force to actively resistant suspects.

50

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Other variations in the policies are also present. Portland places chemical sprays and the
CED together in the middle of the continuum against defensive citizen resistance, while St.
Petersburg places CEDs near the top of the continuum just prior to deadly force and requiring
active resistence. Knoxville instructs officers to use soft hands and pain compliance on verbal
and passive resisters, and then all other forms of force short of deadly on defensive and active
resisters.
Perhaps one of the more interesting agency approaches is found in Colorado Springs.
One of the stated reasons for using a wheel, as opposed to a linear, type model is so officers do
not think about force progression in a ladder or step-by-step format. Clearly, as shown in
Appendix B, the wheel depiction goes so far as to ensure that no force progression is evident
(e.g., lethal force is placed next to chemical agent on one side and soft control techniques on the
other side). Nonetheless, in the text of the policy citizen resistance is laid out in progression
format and explicitly linked to appropriate force options. For example, for soft hand techniques
the policy states “when the subject fails to respond to verbal direction or resists in a defensive
manner,” and for control and compliance techniques (e.g., pain compliance, chemical spray,
CED) the policy states “the subject has become the aggressor, attacks the officer, or there is fear
for the safety of the officer and others” (CSPD GO 705). However, there is also some degree of
uncertainly found within the policy. For example, the policy section dealing directly with CEDs
states this weapon can be deployed when a person “resists detention and arrest, and other
alternatives for controlling them are not reasonable or available under the circumstances” (CSPD
SOP P1-171).

51

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 4-2. Variation in Tactical Placement (Summary Layout)
Agency

Citizen Resistance

Force Response

Columbus

Does not explicitly lay out
resistance to force

Physical Soft, Pain Compliance
Chemical Spray
CED
Physical Hard
Impact (Baton/Flashlight)
Impact (K9)
Impact (Munitions)
Deadly

Charlotte-Mecklenburg

Compliant, Verbal

Officer Presence/Verbal

Passive

Physical Soft, Pain Compliance

Defensive

Chemical Spray

Defensive, Active

Physical Hard

Active, Defensive

CED

Active

Impact

Deadly

Deadly

52

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agency

Citizen Resistance

Force Response

Portland

Compliant

Officer Presence

Verbal

Verbal

Passive

Physical Soft, Pain Compliance

Defensive

Chemical Spray, CED

Active

Physical Hard, Impact (Baton/ASP)

Active

Impact (Munitions)

Deadly

Deadly

Compliant

Officer Presence/Verbal

Verbal, Passive, Defensive

Physical Soft/Hard, Pain, Chemical Spray, CED*

Active

Impact

Deadly

Deadly

Albuquerque

*Policy specifies CED is equivalent to Chemical Spray but cannot
be used on Passive resister

53

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agency

Citizen Resistance

Force Response

Colorado Springs

Compliant

Officer Presence, Verbal

Verbal, Passive

Physical Soft, Pain Compliance

Defensive, Active

Chemical Spray, CED

Active

Physical Hard, Impact

Deadly

Deadly

Compliant, Verbal

Officer Presence, Verbal, Physical Soft

Passive

Pain Compliance

Defensive

Physical Hard, Chemical Spray, Impact

Active

Impact, CED

Deadly

Deadly

St. Petersburg

54

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agency

Citizen Resistance

Force Response

Fort Wayne

Does not explicitly lay out
resistance to force

Officer Presence
Verbal
Physical Soft, Pain Compliance, Chemical Spray
Physical Hard
Impact, CED
Deadly

Knoxville

Compliant

Officer Presence/Verbal

Verbal, Passive

Physical Soft, Pain Compliance

Defensive, Active

Physical Hard, Chemical Spray, CED, Impact

Deadly

Deadly

55

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

There are two additional aspects relevant to force policy that vary. First, in three of the
cities all patrol officers carry a CED (i.e., Portland, Colorado Springs, and Knoxville). In four
cities most, but not all, patrol officers carry a CED (i.e., Columbus, Charlotte-Mecklenburg,
Albuquerque, and St. Petersburg) - meaning that on any given shift there are some officers on
patrol with a CED while other officers do not have a CED. Further, in Fort Wayne, few to none
of the patrol officers have a CED, unless one of the officers is part of the Emergency Response
Team (ERT). Second, while in most of the cities patrol officers themselves are responsible for
documenting their use of force behavior via a force report, two of the cities (Columbus and
Charlotte-Mecklenburg) have a supervisor report system (i.e., a supervisor comes to the scene to
document the force usage).32 In sum, we have eight agencies that vary their force policy
approach in a number of different ways. Below we offer more detail on policy variation for each
agency.

Data Elements
To examine whether certain types of policies offer more beneficial outcomes to police
practitioners we draw on two primary data elements. The first is an officer survey (Chapter 5),
which allows us to examine the extent to which different types of policy approaches provide
officers assistance and guidance with respect to force decision-making. This strategy allows us
to go beyond official policy mandates and administrative rhetoric, and straight to the front-line
workers tasked with carrying out their agency’s policy, to determine if they think it actually
32

All the agencies have some sort of system in place for a supervisor to respond to the scene for severe force
incidents (e.g., cases where a suspect sustained serious or life-threatening injuries). Moreover, CharlotteMecklenburg officials report that supervisor reporting is nearly universal (i.e., supervisor show up to the scene and
do the reporting), while Albuquerque officials report that while “officially” they subscribe to “supervisory
reporting,” in practice patrol officials still document the bulk of the force reporting.

56

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

works. The second set of data elements fall under the broad heading of official data, but is
actually split into three sub-elements in the form of force reports (Chapter 6), citizen complaint
data (Chapter 7), and civil litigation data (Chapter 8). Here we are able to investigate which
policy types are associated with varying degrees of force, injuries to suspects and officers,
citizen complaints, and lawsuits levied for improper force.

57

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 5
Patrol Officer Survey Data

Our patrol officer survey (see Appendix C) was designed to glean information regarding
officer attitudes toward their use of force policy (i.e., assistance, restrictiveness, clarity,
guidance, and review fairness), the impact of the policy on a variety of outcomes (i.e., citizen
injuries, officer injuries, citizen complaints, and lawsuits), and the training associated with the
policy. The survey also captured standard demographic information. The survey was pre-tested
on a sample of current and former Michigan and Florida police officers prior to administering
them to our eight agencies. With the exception of one site33, the patrol officer survey was
administered during organizational roll call sessions, by trained project staff, before the start of
the officers’ shifts. This chapter details the administration of the survey, as well as the patrol
officer responses.

Methodology
Survey Population
We chose to survey a population of police officers assigned to patrol assignments rather
than selecting a sample of officers. This choice was made for a couple of reasons. First, since
we were administering the surveys at roll call meetings, and had ample resources to make visits
to every shift, we aimed to capture the attitudes of all patrol members instead of a select group.
Moreover, given the sensitive nature of several of the survey topics (i.e., the use of force), we did
not want to raise suspicions among officers as to why they were chosen to participate over
33

Fort Wayne Police Department did not use a roll call system, thus we coordinated with the department’s annual inservice training and administered the survey during these training sessions.

58

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

others. In fact, our standard introductory read off to each roll call addressed such concerns by
explaining that, as an organization, officers (individually or collectively) were not selected for
study inclusion for something that they failed to do or for something that they did incorrectly.
Second, since one of our goals was to eventually link attitudinal survey responses with a variety
of behavioral outcomes (e.g., use of force, complaints, lawsuits, etc.), we did not want to
potentially risk missing, through a randomized process, some organizational members that may
be of interest later on (e.g., extremely high or low force users). What follows is a discussion of
the patrol officer survey protocol, as well as the survey results.

Survey Protocol
In administering the officer survey, project staff were generally on-site for a week to ten
days, where the goal was to visit every patrol shift, across each geographic location, at least
twice.34 In order to maximize our survey time while on site, we obtained a master roster of all
sworn personnel from each of our eight agencies. The rosters identified the population of patrol
officers, as well as when and where they were assigned. This information was then matched up
with the daily patrol rosters to ensure that the assignment information was correct. Based on this
information, project staff then prepared a survey plan that would enable them to survey each
shift, across the various precincts/districts, two times. In doing so, project staff mapped
proposed visits around officers’ scheduled days off. For example, if Squad B in District I of
Department X had scheduled days off on Thursday, Friday, and Saturday, survey staff made sure

34

The following represents the dates that project staff surveyed each agency: 9/10/07 - 9/17/07 (Colorado Springs);
9/26/07, 9/28/07 - 9/30/07 (Knoxville); 10/18/07 - 10/31/07 (Charlotte-Mecklenburg); 11/6/07 - 11/7/07, 11/10/07 11/11/07 (Portland); 11/5/07 - 11/10/07, 11/12/ 07 - 11/14/07, 11/16/07 (Fort Wayne); 12/11/07 - 12/17/07
(Albuquerque), 2/17/08 - 2/18/08, 2/20/08 - 2/23/08 (St. Petersburg); and 6/15/08 - 6/23/08 (Columbus).

59

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

to schedule survey times for this squad on Sunday thru Wednesday. In smaller agencies, like St.
Petersburg, this was rather simple, while in departments like Charlotte-Mecklenburg, this was
more arduous. Before going to each site (approximately a week), we obtained up-to-date daily
patrol rosters from the agency to ensure that no “major” changes occurred in terms of officer
assignments. This allowed survey staff to number (and name) each of the surveys beforehand.
Prior to the start of our roll call visits, each of our agency contacts distributed either a
hard copy memorandum or an email of our survey schedule to alert command personnel of the
dates and times of when research staff would be surveying officers. This was done in order to
eliminate any element of “surprise” for shift supervisors, as well as legitimizing the survey with
command staff support.
The amount of time and staff allocated for surveying varied across each of our agencies,
according to size, spatial differentiation (i.e., the number of precincts/divisions), and agency
assistance. For example, the Knoxville Police Department, our smallest and with only two
districts, organized joint roll calls across the districts, enabling us to complete our surveying in
roughly four days. On the other hand, Charlotte-Mecklenburg operated across 13 divisions (with
five to six shifts per division) requiring more survey manpower and approximately 14 days of
surveying. Because of the geographic complexities for two of our sites (Portland and CharlotteMecklenburg), we hired and trained local graduate students to assist in surveying (three from
Portland State University and four from the University of North Carolina at Charlotte,
respectively). These graduate students were all hand picked and recommended from peers of the
project principal investigators. This greatly assisted in efficiently administering the officer
survey across a number of precincts/divisions.

60

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Our roll call visits started by providing attending officers with a brief overview of the
project, highlighting the fact that agencies were chosen based on their divergent use of force
policies and not for something that the department (or individual officers within the department)
failed to do or did incorrectly. Moreover, we explained issues of informed consent,
confidentiality, and anonymity. That is, survey staff explained that officer participation was
voluntary, and if they chose to participate, per federal and university IRB regulations, we needed
their signed consent (see Officer Consent Form in Appendix D). Our introduction also explained
to officers that their responses to the survey were confidential and no one outside of the project
staff would be viewing their individual surveys.
Our confidentiality promises had to be distinguished from the lack of anonymity.
Because we wanted to link attitudinal survey responses with our official data (e.g., use of force,
complaints, lawsuits, etc.), and we wanted to make sure that each officer was accounted for in
the survey process, we needed to identify survey respondents. In doing so, we created a unique
identification number for each officer that was not connected to any official identifying officer
number (e.g., badge number, payroll number, car number, etc). Officers were instructed to alert
us if this happened, and a new identification number would be assigned. Randomly assigned
identification numbers were printed on the last page of the survey, as well as on the consent form
(which was stapled to the front of the survey). On the top of the consent form was the officer’s
name, which was read off in the roll call in order to assure that each officer received a survey.
No names were on the completed officer survey, as consent forms were detached from the survey
once it was handed in to project staff, and officers were instructed that, if they wished, they
could tear off their printed names on the consent forms. Since we had the same identification

61

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

number for each officer on the consent form and the survey, we were able to match up consent
forms with completed surveys.
The patrol officer survey generally took anywhere from ten to twenty-five minutes to
complete. Project staff remained the entire time, in the event that officers had questions about
the survey. Officers were instructed to complete their surveys individually, and not to discuss
any of the items. Those who did not take the survey were either physically absent or declined to
participate. Next, we discuss the response rates across each of our agencies.

Survey Response Rates
Table 5-1 displays our response rates across our eight agencies. Since we were interested
in surveying all police officers with patrol responsibilities, our patrol numbers in the first column
represent the number of officers that were located on the official departmental patrol rosters and
thus assigned to a patrol function. Administrative type assignments, although at the patrol rank,
were not our focus since their likelihood of engaging in the use of force, and thus applying the
less lethal policy, were remote. This column thus represents the number of patrol officers that
were “expected” to be at roll calls when project personnel arrived to conduct the survey.
The second column, in Table 5-1, represents the number of patrol officers that were
physically present to be surveyed during our roll call visits. Even though we had an a priori plan
in place to survey the population of patrol officers, around their scheduled days off, what we
could not control for were those who took an unscheduled day off, attended court, were injured,
on military duty, suspended, or were not present due to some other circumstance. We found
variation across sites between the number of patrol officers that were assigned (on paper) and

62

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

those that were available for each shift. This variation was usually patterned according to
departmental size, and to a lesser extent by spatial differentiation, as there was a greater fit
between column one and column two (i.e., those expected and those present) for the smaller,
more centralized agencies.
The third column, in Table 5-1, depicts the number of patrol officers that participated in
the survey. Subtracting this column from the previous one represents the number of patrol
officers that declined to participate in the survey.
The fourth column, in Table 5-1, indicates the percentage of patrol officers that
participated in our survey relative to the population of officers that we expected at each roll call.
Cumulatively, across all of our eight agencies, we were successful in surveying just over twothirds of the patrol officers that were officially scheduled to be present.
The final column, in Table 5-1, illustrates the percentage of patrol officers that were
present at roll calls and participated in the survey. While there is some minor variation across
sites (ranging from 91.50 in St. Petersburg to 99.08 in Albuquerque), the percentages show that
surveyors were very successful in getting officers to take the survey as long as the officers were
physically present. Cumulatively, surveyors were successful in surveying approximately 96.5
percent of the patrol officers that were present at departmental roll calls (and in-service training
in Fort Wayne). Next, we present the results of patrol officer survey.

63

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-1. Patrol Officer Survey Response Rates
Department
Colorado Springs
Portland
Albuquerque
Fort Wayne
St. Petersburg
Knoxville
Charlotte
Columbus
Total

Patrol #
317
382
456
221
314
176
685
910
3461

@ Roll Call # Surveyed Patrol % Survey %
204
199
62.78
97.55
263
250
65.45
95.06
328
325
71.27
99.08
197
190
85.97
96.45
247
226
71.97
91.50
157
156
88.64
99.36
474
466
68.03
98.31
549
523
57.47
95.26
2419
2335
67.47
96.53

Analyses and Findings
Less Lethal Force Policy Perceptions
A primary component of our survey focused on eliciting officer views of their
departmental less lethal use of force policy. In doing so, we concentrated on asking officers a
series of questions about their overall policy (i.e., assistance, restrictiveness, clarity, guidance,
and review fairness). What follows are the officer responses to each of these questions, all of
which were standard Likert items, enabling the respondent to answer in varying levels of
agreement and disagreement (i.e., “agree strongly,” “agree somewhat,” “disagree somewhat,”
and “disagree strongly”).
We started by asking officers the degree to which their “policy assists officer decision
making” (see Table 5-2). In varying degrees, the majority of respondents, irrespective of agency
and policy type, agreed that overall their policy assisted their decision making. The most
favorable responses were reported from Knoxville (91.6 agreed and 32.5 agreed “strongly”) and
Albuquerque officers (90.9 agreed and 30.4 agreed “strongly”), while St. Petersburg officers
reported the least amount of overall agreement at approximately 74 percent.
64

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-2. Policy Assists Officer Decision Making

Agree Strongly
Agree Somewhat
Disagree Somewhat
Disagree Strongly

Total

CSPD

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

54
(27.3)
110
(55.5)
34
(17.2)
0
(0)

35
(14.3)
160
(65.6)
42
(17.2)
7
(2.9)

98
(30.4)
195
(60.5)
24
(7.5)
5
(1.6)

27
(14.2)
140
(73.7)
22
(11.6)
1
(0.5)

38
(16.8)
129
(57.1)
44
(19.5)
15
(6.6)

50
(32.5)
91
(59.1)
13
(8.4)
0
(0)

103
(22.5)
261
(57.0)
79
(17.2)
15
(3.3)

116
(22.4)
319
(61.5)
76
(14.6)
8
(1.5)

198

244

322

190

226

154

458

519

In assessing policy restrictiveness, we asked officers if they thought the policy was too
restrictive. Table 5-3 displays the results, where there appears to be a good deal of variation
across agencies. The majority of officers from four agencies (i.e., St. Petersburg, 74.6%;
Charlotte-Mecklenburg, 66.4%; Portland, 55.5%; and Columbus, 51.5%) reported that they
agreed that their policy was too restrictive. The majority of the officers from the remaining four
departments disagreed that their policy was too restrictive (i.e., Knoxville, 60.8%; Colorado
Springs, 57.9%; Fort Wayne, 53.7%; and Albuquerque, 51.1%). At the extremes, we found the
least favorable responses among St. Petersburg officers (i.e., the less lethal use of force policy is
too restrictive), while the most favorable responses were found among Knoxville officers.

65

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-3. Policy Too Restrictive

Agree Strongly
Agree Somewhat
Disagree Somewhat
Disagree Strongly

Total

CSPD

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

21
(10.8)
61
(31.3)
72
(36.9)
41
(21.0)

29
(11.7)
108
(43.8)
71
(28.7)
39
(15.8)

33
(10.3)
123
(38.6)
103
(32.3)
60
(18.8)

10
(5.3)
77
(41.0)
71
(37.7)
30
(16.0)

53
(23.6)
115
(51.0)
42
(18.7)
15
(6.7)

8
(5.2)
52
(34.0)
50
(32.7)
43
(28.1)

76
(16.6)
228
(49.8)
109
(23.8)
45
(9.8)

61
(11.9)
203
(39.6)
170
(33.1)
79
(15.4)

195

247

319

188

225

153

458

513

Next, we asked officers the degree to which they agreed that their less lethal force
policy was clear. As Table 5-4 illustrates, the majority of officers, in varying degrees (ranging
from approximately 70% in Portland to roughly 90% in Knoxville) believed that their policy was
clear. Of particular note was the finding that in nearly all agencies (with the exception of
Knoxville) the majority of this agreement was among the “somewhat” category. Once again, the
most favorable less lethal force attitudes were found among Knoxville officers, where
approximately 45 percent of the respondents “agreed strongly” that their policy was clear.
Table 5-4. Policy is Clear

Agree Strongly
Agree Somewhat
Disagree Somewhat
Disagree Strongly

Total

CSPD

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

56
(28.7)
100
(51.3)
32
(16.4)
7
(3.6)

45
(18.4)
125
(51.2)
58
(23.8)
16
(6.6)

112
(35.3)
165
(52.1)
37
(11.7)
3
(0.9)

42
(22.5)
108
(57.7)
34
(18.2)
3
(1.6)

58
(25.8)
115
(51.1)
43
(19.1)
9
(4.0)

70
(45.5)
69
(44.8)
12
(7.8)
3
(1.9)

150
(32.8)
236
(51.7)
60
(13.1)
11
(2.4)

122
(23.6)
286
(55.3)
97
(18.8)
12
(2.3)

195

244

317

187

225

154

457

517

66

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

We also asked officers about the guidance that their policy provided in terms of “when
force can and can’t be used.” As Table 5-5 reports, the majority of officers across all
departments agreed, although in varying degrees, that their policy provides adequate guidance in
terms of when force can and cannot be used. Also, like the previous guidance question, we
found Knoxville officers to be the most favorable (approximately 44% agreed, and 48% agreed
strongly), while Portland officers were the least favorable of the eight agencies (approximately
52% agreed, and only 18% agreed strongly). Across all departments, with the exception of
Knoxville, the modal response was “agree somewhat.”
Table 5-5. Policy Provides Adequate Guidance in Terms of
When Force Can and Can’t Be Used

Agree Strongly
Agree Somewhat
Disagree Somewhat
Disagree Strongly

Total

CSPD

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

49
(24.9)
112
(56.9)
32
(16.2)
4
(2.0)

45
(18.3)
129
(52.4)
61
(24.8)
11
(4.5)

112
(34.9)
178
(55.4)
26
(8.1)
5
(1.6)

34
(18.0)
120
(63.5)
32
(16.9)
3
(1.6)

62
(27.5)
126
(56.0)
29
(12.9)
8
(3.6)

74
(48.0)
68
(44.2)
10
(6.5)
2
(1.3)

162
(35.4)
242
(52.8)
49
(10.7)
5
(1.1)

123
(23.7)
319
(61.5)
70
(13.5)
7
(1.3)

197

246

321

189

225

154

458

519

The final policy assessment survey item focused on supervisory review and use of force
reporting. This question asked officers to assess the extent to which they agreed (or disagreed)
that their “policy regarding supervisory of use of force reports is fair.” As Table 5-6 indicates,
the majority of officers, across all agencies, agreed that the supervisory review of use of force
reports is fair. We did find variation in this majority agreement, ranging from 75 percent overall
agreement in Colorado Springs to 95.4 percent in Portland. In terms of intensity of agreement,

67

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Knoxville officers reported the highest percentage of “agree strongly” at 37 percent. While we
found high overall agreement, across all agencies, the bulk of this agreement (and modal
response) was among the “agree somewhat” response category.
Table 5-6. Policy Regarding Supervisory Review is Fair
CSPD
40
(20.3)
Agree Somewhat
108
(54.8)
Disagree Somewhat 40
(20.3)
Disagree Strongly
9
(4.6)
Agree Strongly

Total

197

PPB
39
(16.1)
153
(63.2)
38
(15.7)
12
(5.0)

APD
68
(21.3)
194
(60.5)
52
(16.3)
6
(1.9)

FWPD
32
(17.1)
126
(67.4)
22
(11.8)
7
(3.7)

SPPD
45
(20.0)
125
(55.6)
45
(20.0)
10
(4.4)

KPD
57
(37.0)
86
(55.9)
8
(5.2)
3
(1.9)

CMPD
119
(26.1)
257
(56.3)
72
(15.8)
8
(1.8)

CPD
109
(21.1)
328
(63.5)
71
(13.7)
9
(1.7)

242

320

187

225

154

456

517

Summary
The aim of the first section, reported here, of the patrol officer survey was to assess
officers’ perception of their less lethal use of force policy. Our primary topics focused on
overall policy assistance, restrictiveness, clarity, guidance, and review fairness.
Across the various facets of attitudes toward respective use of force policies, we find
three patterns that emerge. First, although we found variation in responses within and across
agencies, patrol officers generally expressed favorable views of their less lethal use of force
policy.
Second, one agency consistently stood out in terms of patrol officers positive attitudes
toward their use of force policy. That department was Knoxville, where officers reported the
positive responses of their policy (at higher percentages) in nearly every table found in this
section (see tables 5-2 through 5-6).

68

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Third, a few departments (i.e., St. Petersburg, Portland, and Colorado Springs)
consistently distinguished themselves by either reporting the lowest percentage of favorable
responses or by responding unfavorably. For example, St. Petersburg officers reported the
lowest percentage of agreement, across the eight sites, in terms of the policy providing assistance
in general (see Table 5-2). These officers also expressed greater concerns (i.e., majority
agreement) that the policy was too restrictive (see Table 5-3). Portland officers, on the other
hand, noted the least amount of agreement that their policy was clear (see Table 5-4). The same
patterns were found for Portland officers with respect to when force can and cannot be used (see
Table 5-5). Finally, Colorado Springs officers distinguished themselves by reporting the least
overall agreement that the supervisory review of their use of force is fair (see Table 5-6).

Outcomes of Less Lethal Force Policy
The next section of our patrol officer survey queried respondents on a variety of
“outcomes” of their respective force policies. More specifically, we asked officers to assess the
impact that their less lethal use of force policy had on potential citizen injuries, officer injuries,
citizen complaints, and lawsuits. For all items, we asked officers to respond with one of the
following: “increases the likelihood,” “decreases the likelihood,” or “neither increases nor
decreases the likelihood.”
Table 5-7 displays the results of patrol officers’ perceptions of the effect of their less
lethal use of force policy on potential citizen injuries. The majority of officers, across all
agencies, reported that their policy decreases the likelihood of potential citizen injuries, ranging
from 50.3 percent in Fort Wayne to 69.7 percent in Knoxville. Interestingly, while most

69

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

respondents did not report that their policy increased the likelihood of citizen injuries, over 10
percent of Fort Wayne officers (11.6) did. This contrasts dramatically with Knoxville, where
less than one percent (0.6) perceived that their use of force policy increased the likelihood of
injuries to citizens.
Table 5-7. Effect of Policy on Potential Citizen Injuries
Increases the
Likelihood

CSPD
14
(7.0)

PPB
13
(5.3)

APD
12
(3.8)

FWPD
22
(11.6)

SPPD
17
(7.7)

KPD
1
(0.6)

CMPD
19
(4.2)

CPD
8
(1.6)

Decreases the
Likelihood

120
(60.3)

142
(58.2)

198
(61.8)

95
(50.3)

147
(66.2)

108
(69.7)

271
(59.4)

295
(57.3)

Neither Increases
Nor Decreases the
Likelihood

65
(32.7)

89
(36.5)

110
(34.4)

72
(38.1)

58
(26.1)

46
(29.7)

166
(36.4)

211
(41.1)

199

244

320

189

222

155

456

514

Total

Turning next to the impact of respondents’ use of force policy on patrol officer injuries,
we found contrasting results from that reported for citizen injuries (see Table 5-8). Among all
agencies, we find much more variation across likelihood response categories (“increases,”
“decreases,” or “neither increases or decreases”). At the higher ends, St. Petersburg and
Charlotte-Mecklenburg officers reported that their policy increases the likelihood of officer
injuries (45.7% and 44.2% respectively). Knoxville officers reported the least percentage
(19.4%) of officer injury potential, which was still higher than any agency reported for potential
citizen injuries. Finally, Knoxville officers were the only group where the majority (58%)
reported that their policy actually decreases the likelihood of officer injuries.

70

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-8. Effect of Policy on Potential Officer Injuries
Increases the
Likelihood

CSPD
69
(34.7)

PPB
88
(36.1)

APD
88
(27.5)

FWPD
61
(32.3)

SPPD
101
(45.7)

KPD
30
(19.4)

CMPD
201
(44.2)

CPD
143
(27.8)

Decreases the
Likelihood

96
(48.2)

98
(40.1)

144
(45.0)

69
(36.5)

84
(38.0)

90
(58.0)

138
(30.3)

216
(42.0)

Neither Increases
Nor Decreases the
Likelihood

34
(17.1)

58
(23.8)

88
(27.5)

59
(31.2)

36
(16.3)

35
(22.6)

116
(25.5)

155
(30.2)

199

244

320

189

221

155

455

514

Total

Our third assessment of policy outcomes dealt with potential citizen complaints. Again,
like that found in assessing potential officer injuries, we found (see Table 5-9) a good deal of
variation across response categories. Of interest was the fact that officers in six agencies (i.e.,
Colorado Springs, Portland, Albuquerque, Fort Wayne, St. Petersburg, and Columbus) reported
higher percentages that their policy actually increases the likelihood of citizen complaints over
decreases the likelihood. Knoxville and Charlotte-Mecklenburg officers reported higher
percentages (37.4 and 33.2 respectively) that the policy decreases the likelihood over increases
the likelihood. Finally, over a third of all respondents (ranging from 33.5% in St. Petersburg to
42.6% in Knoxville) reported that their policy did not increase or decrease potential citizen
complaints.

71

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-9. Effect of Policy on Potential Citizen Complaints
Increases the
Likelihood

CSPD
79
(39.7)

PPB
97
(39.8)

APD
121
(37.8)

FWPD
67
(35.4)

SPPD
81
(36.6)

KPD
31
(20.0)

CMPD
135
(29.7)

CPD
209
(40.6)

Decreases the
Likelihood

44
(22.1)

49
(20.1)

75
(23.4)

43
(22.8)

66
(29.9)

58
(37.4)

151
(33.2)

93
(18.1)

Neither Increases
Nor Decreases the
Likelihood

76
(38.2)

98
(40.1)

124
(38.8)

79
(41.8)

74
(33.5)

66
(42.6)

169
(37.1)

213
(41.3)

199

244

320

189

221

155

455

515

Total

The final outcome item assesses officers’ perceptions of the effect of the less lethal force
policy on potential lawsuits. As Table 5-10 illustrates, the most favorable assessments of the
force policy were found among Knoxville officers, as nearly half of the officers (49.4%) reported
that the policy decreases the likelihood of potential lawsuits. The least favorable assessments of
the policy was found among Portland respondents, who reported the highest percentage (31.3)
that the policy increases potential lawsuits, and the lowest percentage (22.2) that the policy
decreases the likelihood potential lawsuits. Of note, with the exception of St. Petersburg and
Knoxville, across all of the agencies, the modal response was that the policy “neither increases
nor decreases the likelihood” of potential lawsuits.

72

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-10. Effect of Policy on Potential Lawsuits
Increases the
Likelihood

CSPD
42
(21.2)

PPB
76
(31.3)

APD
91
(28.4)

FWPD
55
(29.1)

SPPD
45
(20.3)

KPD
7
(4.5)

CMPD
72
(15.8)

CPD
120
(23.3)

Decreases the
Likelihood

66
(33.3)

54
(22.2)

95
(29.7)

63
(33.3)

91
(41.0)

76
(49.4)

184
(40.3)

169
(32.8)

Neither Increases
Nor Decreases the
Likelihood

90
(45.5)

113
(46.5)

134
(41.9)

71
(37.6)

86
(38.7)

71
(46.1)

200
(43.9)

226
(43.9)

198

243

320

189

222

154

456

515

Total

Summary
The focus of this section of the patrol officer survey was to assess respondents’ views of
the potential impact of the less lethal use of force policy on a variety of important outcomes.
That is, we asked officers if they thought that their respective policy increases, decreases, or
neither increases nor decreases the likelihood of potential injuries to citizens and officers, citizen
complains, and lawsuits.
Across each of these outcomes, there were a few findings that stood out. First, Knoxville
respondents, as a group, once again expressed the most favorable assessments of their policy, as
they reported the highest percentage of responses that their policy decreases the likelihood of
potential citizen and officer injuries, citizen complaints, and potential lawsuits.
Second, across the outcomes, we found some patterns that emerged, especially with
respect to the degree of variation within and across agencies. In terms of the likelihood of
potential citizen injuries, the majority of respondents in all departments believed that their less
lethal force policy decreased the likelihood, although Fort Wayne comprised the fewest
percentage of officers in this response category.

73

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

By contrast, with respect to potential officer injuries, we found more division within and
across agencies, although in every department respondents reported more of a likelihood for
increases in officer injuries compared to citizen injuries. While the modal response, across
agencies, was that the policy “decreases the likelihood for officer injuries (except for St.
Petersburg and Charlotte-Mecklenburg where the mode was “increases”), Knoxville was the
only agency where the majority of officers were found in this response category.
For potential citizen complaints, we found even more variation in responses, although
(with the exception of Knoxville) approximately 30 percent (in Charlotte-Mecklenburg) to 40
percent (in Columbus) of the respondents thought their policy “increases the likelihood” of such
actions, while roughly the same percentages reported “neither increases nor decreases the
likelihood.”
Finally, for potential lawsuits, the modal response across agencies was either “decreases
the likelihood” (i.e., Knoxville and St. Petersburg) or “neither increases nor decreases the
likelihood” (i.e., all others). Compared to all other agencies, Portland comprised the highest
percentage of officers (31.3%) that believed that their policy “increases the likelihood” for
potential lawsuits.

Training and Less Lethal Force Policy
Next, we asked respondents to assess the adequacy of their agency’s less lethal academy
and in-service training. Both questions utilized Likert questionnaire survey items with response
categories of “agree strongly,” “agree somewhat,” “disagree somewhat,” and “disagree
strongly.” This section also allowed officers to report areas of academy and in-service training
that they believed could receive more attention. As Table 5-11 illustrates, the majority of
74

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

officers, in varying degrees, agreed that their pre-service less lethal force training was adequate.
Knoxville and Albuquerque officers were the most favorable (91% and 90.4%, respectively),
while St. Petersburg and Colorado Springs officers were at the lower end of agreement (62.8%
and 65.9%, respectively). While assessments of pre-service less lethal force training, across all
agencies, were more positive than negative, it is important to note that the highest percentage of
agreement (modal response) was found within the “agree somewhat” category.
Table 5-11. Pre-Service LTLF Training is Adequate
CSPD
32
(16.2)

PPB
55
(23.2)

APD
115
(35.7)

FWPD
46
(24.6)

SPPD
35
(16.1)

KPD
55
(35.5)

CMPD
159
(34.9)

CPD
177
(34.3)

Agree Somewhat

98
(49.7)

103
(43.5)

176
(54.7)

100
(53.5)

102
(46.7)

86
(55.5)

212
(46.6)

270
(52.4)

Disagree Somewhat

47
(23.9)

46
(19.4)

21
(6.5)

32
(17.1)

51
(23.4)

14
(9.0)

71
(15.6)

60
(11.6)

Disagree Strongly

20
(10.2)

33
(13.9)

10
(3.1)

9
(4.8)

30
(13.8)

0
(0)

13
(2.9)

9
(1.7)

197

237

322

187

218

155

455

516

Agree Strongly

Total

In addition to asking about the adequacy of pre-service training on less lethal force, we
also asked officers, in an open-ended format, if any part of the training should receive more
attention. Based on these responses, we coded the following twelve categories: more/continued
training, defensive tactics/controlling techniques, ground fighting/take downs, hand to
hand/hands on, any weapons or deadly force, role playing/scenarios/practical applications,
dealing with verbal or passively resistant citizens, dealing with actively resistant citizens, aspects
of the policy or use of force documentation, officer verbal skills or verbal judo, all aspects of
force, and “other” (e.g., officer survival, situational awareness, decision making, disarming
suspects, when to use or not use force, etc.). Roughly one-fifth to two-fifths of the officers that
75

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

responded to the adequacy of the pre-service training (in Table 5-11 above), followed up by
describing the parts of the training that should receive more attention (see Table 5-12).
Looking across the three agencies (i.e., St. Petersburg, Colorado Springs, and Portland)
that expressed the least overall agreement that the pre-service training was adequate, we found
that Colorado Springs and Portland both (among other things) believed that more attention
should be paid to “hand to hand/hands on” (25% in Colorado Springs and 14.1% in Portland)
and “weapons/deadly force” (20.2% in Colorado Springs and 15.6% in Portland). St. Petersburg
officers reported that “more/continued” training was needed (14.5%), as well as “defensive
tactics/control techniques” (14.5%) and “scenarios/role playing/practical applications” (17.7%)
instruction. Interestingly, none of the Knoxville officers, who expressed the most favorable
attitudes toward their pre-service training, stated that “more/continued” training was needed,
while 41.4 percent of the officers that did the follow up question noted that more “hand to
hand/hands on” techniques should be conducted.

76

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-12. Pre-Service LTLF Training that Should Receive More Attention

More/Continued

CSPD
4
(4.8)

PPB
7
(10.9)

APD
13
(13.7)

FWPD
12
(25.0)

SPPD
9
(14.5)

KPD
0
(0)

CMPD
7
(5.3)

CPD
1
(1.0)

Defensive Tactics/
Control Techniques

13
(15.5)

6
(9.4)

8
(8.4)

4
(8.3)

9
(14.5)

4
(13.8)

20
(15.2)

4
(4.1)

4
(4.8)

9
(14.1)

7
(7.4)

1
(2.1)

0
(0)

1
(3.4)

4
(3.0)

15
(15.5)

Hand to Hand/
Hands On

21
(25.0)

9
(14.1)

13
(13.7)

4
(8.3)

8
(12.9)

12
(41.4)

18
(13.6)

24
(24.7)

Weapons/
Deadly Force

17
(20.2)

10
(15.6)

12
(12.6)

9
(18.8)

4
(6.5)

2
(6.9)

9
(6.8)

1
(1.0)

Scenarios/Role
Playing/Practical
Applications

6
(7.1)

5
(7.8)

12
(12.6)

2
(4.2)

11
(17.7)

1
(3.4)

22
(16.7)

11
(11.3)

Verbal/Passive
Resistant
Citizens

5
(6.0)

4
(6.3)

3
(3.2)

2
(4.2)

2
(3.2)

0
(0)

11
(8.3)

12
(12.4)

Active Resistant
Citizens

1
(1.2)

2
(3.1)

2
(2.1)

0
(0)

0
(0)

2
(6.9)

2
(1.5)

2
(2.1)

Policy/Force
Documentation

0
(0)

3
(4.7)

5
(5.3)

4
(8.3)

4
(6.5)

0
(0)

3
(2.3)

4
(4.1)

Verbal Skills

0
(0)

2
(3.1)

5
(5.3)

0
(0)

0
(0)

1
(3.4)

3
(2.3)

2
(2.1)

All Aspects

1
(1.2)

2
(3.1)

2
(2.1)

3
(6.3)

4
(6.5)

3
(10.3)

9
(6.8)

4
(4.1)

Other

12
(14.3)

6
(9.4)

13
(13.7)

7
(14.6)

11
(17.7)

3
(10.3)

24
(18.2)

17
(17.5)

Total

84

95

48

62

29

132

97

Ground Fighting/
Take Downs

64

The last two questions in Section IV of the survey focused on perceptions of inservice less lethal force training. Like that noted for pre-service training, we found variation in
agreement across the eight agencies. As reported in Table 5-13, the lowest percentages of
agreement were found among Fort Wayne (56.1% agreed, and only 13.8% agreed “strongly”)

77

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

and Colorado Springs (59.4% agreed, and only 14.2% agreed “strongly”) respondents. On the
other end of the spectrum were Columbus officers, where 88.5 percent of the officers agreed that
their in-service training was adequate.
Table 5-13. In-Service LTLF Training is Adequate
CSPD
28
(14.2)
Agree Somewhat
89
(45.2)
Disagree Somewhat 54
(27.4)
Disagree Strongly
26
(13.2)
Agree Strongly

Total

197

PPB
55
(22.7)
113
(46.7)
53
(21.9)
21
(8.7)

APD
94
(29.3)
157
(48.9)
51
(15.9)
19
(5.9)

FWPD
26
(13.8)
80
(42.3)
55
(29.1)
28
(14.8)

SPPD
48
(21.7)
92
(41.7)
50
(22.6)
31
(14.0)

KPD
54
(34.8)
77
(49.7)
20
(12.9)
4
(2.6)

CMPD
145
(31.9)
215
(47.2)
65
(14.3)
30
(6.6)

CPD
161
(31.4)
293
(57.1)
47
(9.2)
12
(2.3)

242

321

189

221

155

455

513

When asked to provide information on the areas of in-service training that should receive
more attention (see Table 5-14), 23.5 percent of Fort Wayne and 27 percent of Colorado Springs
officers reported that they believed “more/continued training” should be conducted. A similar
sentiment was also noted, not just among the least favorable in-service training assessment
agencies, but also among Albuquerque (25.9%), St. Petersburg (41.1%), and CharlotteMecklenburg (32.1%) respondents. Among the most favorable perceptions of in-service less
lethal force training, we found that roughly one-fifth of Columbus officers requested more
attention on “ground fighting/take downs,” while 30.4 percent of Knoxville respondents
requested more “hand to hand/hands on” training (which was similar to what they requested for
pre-service training curriculums).

78

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-14. In-Service LTLF Training that Should Receive More Attention

More/Continued

CSPD
24
(27.0)

PPB
15
(19.2)

APD
21
(25.9)

FWPD
16
(23.5)

SPPD
23
(41.1)

KPD
2
(8.7)

CMPD
36
(32.1)

CPD
12
(15.6)

Defensive Tactics/
Control Techniques

15
(16.9)

19
(24.4)

6
(7.4)

10
(14.7)

8
(14.3)

5
(21.7)

14
(12.5)

6
(7.8)

6
(6.7)

5
(6.4)

3
(3.7)

0
(0)

0
(0)

0
(0)

1
(0.9)

16
(20.8)

Hand to Hand/
Hands On

11
(12.4)

7
(9.0)

9
(11.1)

7
(10.3)

3
(5.4)

7
(30.4)

14
(12.5)

7
(9.1)

Weapons/
Deadly Force

6
(6.7)

7
(9.0)

13
(16.0)

12
(17.6)

5
(8.9)

2
(8.7)

9
(8.0)

0
(0)

Scenarios/Role
Playing/Practical
Applications

5
(5.6)

7
(9.0)

2
(2.5)

6
(8.8)

6
(10.7)

0
(0)

6
(5.4)

7
(9.1)

Verbal/Passive
Resistant
Citizens

2
(2.2)

2
(2.6)

3
(3.7)

2
(2.9)

0
(0)

0
(0)

6
(5.4)

8
(10.4)

Active Resistant
Citizens

2
(2.2)

2
(2.6)

0
(0.0)

0
(0)

0
(0)

0
(0)

2
(1.8)

2
(2.6)

Policy/Force
Documentation

1
(1.1)

3
(3.8)

3
(3.7)

1
(1.5)

3
(5.4)

0
(0)

2
(1.8)

3
(3.9)

Verbal Skills

0
(0)

1
(1.3)

2
(2.5)

0
(0)

0
(0)

1
(4.3)

3
(2.7)

1
(1.3)

All Aspects

5
(5.6)

1
(1.3)

1
(1.2)

5
(7.4)

3
(5.4)

3
(13.0)

4
(3.6)

3
(3.9)

Other

12
(13.5)

9
(11.5)

18
(22.2)

9
(13.2)

5
(8.9)

3
(13.0)

15
(13.4)

12
(15.6)

Total

89

78

81

68

23

112

77

Ground Fighting/
Take Downs

56

Summary
The goal of this section of the survey was to assess officers’ perceptions of the adequacy
of their pre-service and in-service less lethal use of force training. While the majority of
respondents, from all agencies, agreed that both types of training were adequate, we did find
79

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

variation within and among departments. First, assessments of pre-service less lethal use of
force training were generally more favorable than in-service training. In terms of pre-service
training, Knoxville and Albuquerque officers were the most favorable, while St. Petersburg,
Colorado Springs, and Portland officers were the least favorable. For in-service training, the
most favorable responses were reported among Columbus officers, while the least favorable
were noted for Fort Wayne and Colorado Springs officers.
A subset of respondents answered our open-ended questions regarding areas of both
types of training that should receive more attention. Across the agencies with the least favorable
views of the pre-service training, a good deal of St. Petersburg officers, among other things,
wanted more attention paid to “defensive tactics/control techniques,” “scenarios/role
playing/practical applications,” and “more/continued” training, while both Colorado Springs and
Portland officers generally wanted more “hand to hand/hands on” and “weapons/deadly force”
training. In terms of the agencies with the least favorable attitudes toward in-service force
training, among other areas, a fair number of both Fort Wayne and Colorado Springs officers
wanted “more/continued training” (as did many other respondents) as well as “defensive
tactics/control techniques” instruction.

Background and Demographics
In this final section, we report findings on a variety of respondent background and
demographic factors, as well as future aspirations. While the general patterns (i.e., majority) of
these factors were somewhat similar across agencies, we did find fair amount of variation as
well. As Table 5-15 displays, over three quarters of our respondents from all agencies were

80

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

males - ranging from 93.2 percent in Fort Wayne to 81 percent in St. Petersburg. For both sex
and race (below), given potential confidentiality concerns by officers, we chose not to ask
respondents on the survey. Instead, in constructing these measures, we utilized official agency
records (as part of our master departmental rosters), linking that data with our unique
identification number for each officer.
In terms of race, the majority of all respondents were white, although we do find
variation across agencies. With respect to non-White respondents, the fewest were found in
Knoxville (4.4%), while the most were found in Albuquerque (39.1%). Moreover, in five of our
cities, the modal non-white respondent was Black. The modal non-White respondent in
Colorado Springs and Albuquerque was Hispanic, while in Portland the modal non-White
respondent was Asian.
We also asked officers to report their current marital status. With the exception of St.
Petersburg, the majority of officers, from all agencies, were married (ranging from 79.8% in
Portland to 61.7% in Albuquerque). Overall, roughly one-tenth of the respondents reported that
they were divorced - ranging from 8.5 percent in Knoxville to 14.1 percent in Colorado Springs.
In an effort to possibly later deduce socializing influences, officers were asked to inform
us as to whether they had a family member that was also in the policing occupation. The
majority of respondents, across all sites, did not have a family member that was a police officer.
The highest percentage of respondents that did have one (or more) was found in Albuquerque
(38.2%), while the fewest percentage was noted for Charlotte-Mecklenburg (30%).
In terms of military experience, the majority of all respondents (regardless of
department), did not have any military background. We did find variation in such experience

81

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

across sites, as 44.2 percent of Colorado Springs officers reported a military background, while
only 22.9 percent of Portland officers reported such experience. We followed this question up,
like we did later for college experience, by asking respondents (who responded “yes”) if they
believed that their military experience helped prepare them to be a police officer. With the slight
exception of Knoxville (76.7% responded “yes”), respondents overwhelmingly reported that they
believed that their prior military experience helped them prepare to be a police officer (ranging
from 98.3% in Fort Wayne to 85.5% in Portland).
With respect to education, we asked officers to report their highest level of formal
education that they completed (i.e., from “less than high school” to “graduate degree”).
Interestingly, the vast majority of respondents had some form of college experience (ranging
from 79.8% in Albuquerque to 99.2% in Portland). Across this college experience though, we
found variation in levels of completion. For example, the majority of college experience in
Knoxville, St. Petersburg, and Albuquerque is found at the two years or more of college (without
a bachelors degree) and below (56.6%, 53.4%, and 53.4%, respectively). By contrast, 60.6
percent of Portland, 50 percent of Charlotte-Mecklenburg, and 45.9 percent of Colorado Springs
officers completed a bachelors degree. We also find that graduate work (and degrees) was much
less common among respondents in all agencies (ranging from 3.7% in Columbus to 13.8% in
Portland). As evidenced by reported bachelors degrees, graduate work, and graduate degrees,
Portland respondents were the highest educated patrol officers, while the opposite was true of
Knoxville and Albuquerque patrol officers.
As we did for military background, we asked officers if they believed that their college
experiences (across all levels from “some” college to graduate degree) helped prepare them for

82

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

their occupation as a police officer. Of interest is the fact that we did not find the same degree of
support for college experience helpfulness in policing as we did for those reporting military
experience. Among those with college experience, we found in three agencies only, Colorado
Springs, Portland, and Fort Wayne, where the majority of respondents believed college work
help prepare them to be a police officer. Interestingly, among the two departments with the
fewest percentages of officers with college experience (i.e., Albuquerque and Knoxville) we
found the fewest percentages of respondents (40.4% and 32.3%, respectively) that responded
“yes” that their college experience prepared them for policing.
Our final three measures (i.e., age, organizational experience, and overall policing
experience), like that for sex and race reported previously, were not asked of respondents, but
instead gleaned from master rosters from each site.35 In terms of age of respondents, we found
some degree of variation across sites. Our oldest respondents (mean of approximately 38 years
old and a median of 36) were found among Colorado Springs and Portland officers. Average age
across the rest of our sites ranged from a mean of 34.3 years old in Albuquerque to 36.7 years
old in Fort Wayne.
With respect to organizational tenure, the average patrol officer experience ranged from
5.1 years in Albuquerque to 9.1 years in Fort Wayne. The officers with the least amount of
organizational tenure were found in Albuquerque (mean = 5.1, median = 4.0), Knoxville (mean =
6.9, median = 5.0), and St. Petersburg (mean = 7.5, median = 4.0). Our survey also asked if
35

We were able to obtain, in many instances, this same type of information for education, but found that most
agencies did not keep up-to-date, or complete, records for officer education. That is, departments would only record
college degrees (and not college experience with less than a degree) at the time the officer was hired. As such, any
college work done that did not result in a degree, or education while serving as a police officer was usually not
accounted for in departmental records (or updated infrequently). Because of these concerns, we chose to capture this
data, with the level and rigor we deemed as appropriate, on the survey. The same was not true for other officer-level
data that we obtained from our agencies, such as sex, race, date of birth, and hire date, which were generally
extremely accurate.

83

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

respondents had prior police experience with another police agency, and if so, how much. We
added this time to the organizational tenure to capture overall police experience. This allowed
us to account for instances where an officer transfers with 15 years of experience to one of our
sites, where s/he might have worked for one year, and reports that information to us. While
his/her organizational tenure is accurately captured at one year, this officer’s total experience is
missed, and thus they may not be thinking and processing police information attitudinally like
someone with one year of experience. This measure allowed us to account for such
discrepancies. As Table 5-15 illustrates, in comparing organizational experience to overall
experience, the average years of police experience increased across agencies. The results reveal
similar trends as that noted for organizational experience. That is, the least experienced officers
were still found among Albuquerque (mean = 6.4, median = 4.0), Knoxville (mean = 7.9, median
= 7.0), and St. Petersburg (mean = 8.0, median = 4.0) respondents, while the most experienced
officers were found among Fort Wayne (mean = 10.3, median = 9.0), Portland (mean = 10.3,
median = 9.0), and Colorado Springs (mean = 10.4, median = 8.0) respondents.

84

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-15. Background and Demographics
CSPD

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

21
(10.6)
178
(89.4)

31
(12.8)
212
(87.2)

36
(11.4)
281
(88.6)

13
(6.8)
177
(93.2)

43
(19.0)
183
(81.0)

13
(8.3)
143
(91.7)

56
(12.0)
410
(88.0)

57
(10.9)
466
(89.1)

159
(79.9)
11
(5.5)
20
(10.1)
7
(3.5)
2
(1.0)
0
(0)

213
(87.7)
10
(4.1)
7
(2.9)
12
(4.9)
1
(0.4)
0
(0)

192
(60.9)
7
(2.2)
113
(35.9)
3
(1.0)
0
(0)
0
(0)

162
(85.2)
21
(11.1)
6
(3.2)
1
(0.5)
0
(0)
0
(0)

179
(79.2)
24
(10.6)
17
(7.5)
0
(0)
0
(0)
6
(2.7)

149
(95.6)
4
(2.6)
0
(0)
1
(0.6)
1
(0.6)
1
(0.6)

370
(79.4)
74
(15.9)
11
(2.4)
10
(2.1)
1
(0.2)
0
(0)

469
(89.6)
47
(9.0)
1
(0.2)
1
(0.2)
0
(0)
5
(1.0)

28
(14.1)
138
(69.8)
4
(2.0)
28
(14.1)
0
(0)

26
(10.7)
193
(79.8)
1
(0.4)
22
(9.1)
0
(0)

82
(25.5)
198
(61.7)
5
(1.6)
35
(10.9)
1
(0.3)

20
(10.6)
144
(76.2)
4
(2.1)
20
(10.6)
1
(0.5)

91
(41.0)
105
(47.2)
1
(0.5)
25
(11.3)
0
(0)

43
(28.1)
95
(62.1)
2
(1.3)
13
(8.5)
0
(0)

113
(24.8)
288
(63.2)
11
(2.4)
44
(9.6)
0
(0)

106
(20.6)
354
(68.7)
6
(1.2)
49
(9.5)
0
(0)

135
(68.2)
63
(31.8)

165
(67.3)
80
(32.7)

197
(61.8)
122
(38.2)

126
(66.3)
64
(33.7)

146
(65.8)
76
(34.2)

103
(66.5)
52
(33.5)

324
(70.0)
139
(30.0)

328
(63.2)
191
(36.8)

111
(55.8)
88
(44.2)

189
(77.1)
56
(22.9)

216
(66.9)
107
(33.1)

132
(69.5)
58
(30.5)

167
(74.9)
56
(25.1)

110
(71.9)
43
(28.1)

298
(64.5)
164
(35.5)

367
(70.8)
151
(29.2)

5
(5.7)
83
(94.3)

8
(14.5)
47
(85.5)

9
(8.4)
98
(91.6)

1
(1.7)
57
(98.3)

7
(12.7)
48
(87.3)

10
(23.3)
33
(76.7)

18
(11.1)
144
(88.9)

21
(13.9)
130
(86.1)

Sex
Female
Male
Race
White
Black
Hispanic
Asian
Native American
Other
Marital Status
Single
Married
Separated
Divorced
Widowed
Family Officers
No
Yes
Military Experience
No
Yes
Help as Officer
No
Yes

85

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CSPD
Education
< High School

0
(0)
H.S./GED Diploma 3
(1.5)
Some Jr. College
14
(7.1)
Associates Degree 35
(17.7)
>2 Yrs. College
35
(17.7)
Bachelors Degree 91
(45.9)
Graduate School
12
(6.1)
Graduate Degree
8
(4.0)

College Help as
Officer
No
Yes

87
(45.1)
106
(54.9)

PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

1
(0.4)
1
(0.4)
16
(6.5)
13
(5.3)
32
(13.0)
149
(60.6)
18
(7.3)
16
(6.5)

0
(0)
64
(20.2)
100
(31.4)
24
(7.5)
46
(14.5)
61
(19.2)
17
(5.3)
6
(1.9)

0
(0)
23
(12.2)
36
(19.0)
24
(12.7)
32
(16.9)
65
(34.5)
5
(2.6)
4
(2.1)

0
(0)
7
(3.2)
49
(22.2)
37
(16.7)
32
(14.5)
81
(36.6)
8
(3.6)
7
(3.2)

0
(0)
29
(19.1)
54
(35.5)
13
(8.6)
19
(12.5)
28
(18.4)
2
(1.3)
7
(4.6)

1
(0.2)
46
(10.1)
73
(16.0)
57
(12.5)
21
(4.6)
228
(50.0)
13
(2.9)
17
(3.7)

0
(0)
81
(15.7)
106
(20.6)
73
(14.2)
72
(14.0)
164
(31.8)
12
(2.3)
7
(1.4)

102
(41.6)
143
(58.4)

155
(59.6)
105
(40.4)

73
(44.0)
93
(56.0)

110
(51.4)
104
(48.6)

86
(67.7)
41
(32.3)

219
(53.9)
187
(46.1)

247
(58.4)
176
(41.6)

Age
Mean
Median
SD
N
Range

37.7
36.0
8.4
199
25-59

37.5
36.0
7.8
241
24-61

34.3
33.0
9.1
311
21-70

36.7
35.5
7.0
190
24-60

34.6
33.0
9.0
226
21-61

34.9
34.0
8.2
156
22-60

34.7
34.0
7.2
466
22-63

35.2
35.0
7.4
523
22-61

Experience (years)
Organization
Mean
Median
SD
N
Range

8.7
7.0
7.4
199
0-33

8.7
8.0
5.9
242
1-31

5.1
4.0
5.2
316
0-34

9.1
7.0
6.9
190
0-31

7.5
4.0
8.4
226
0-38

6.9
5.0
5.8
156
1-32

7.8
7.0
6.2
466
0-28

7.8
7.0
6.3
523
0-36

10.4
8.0
7.9
199
0-33

10.3
9.0
6.8
242
1-32

6.4
4.0
5.8
316
0-38

10.3
9.0
6.8
190
2-31

8.0
4.0
8.5
226
0-38

7.9
7.0
6.2
156
1-32

8.6
8.0
6.4
466
0-30

8.5
7.0
6.6
523
0-36

Overall
Mean
Median
SD
N
Range

86

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The final table, 5-16, assesses respondents’ occupational aspirations in terms of the
relative importance they placed on being promoted to a higher rank and specialized unit, as well
as their expectations of their highest ranking level when they retire from policing. For the first
two questions, regarding the importance of getting promoted vertically and laterally, we utilized
standard Likert survey items where respondents answered in terms of relative importance (i.e.,
“very unimportant,” “somewhat unimportant,” “somewhat important,” and “very important”). In
terms of promotion to a higher rank, respondents were generally split in agreement, across sites,
as to its overall importance. While the modal response, for every site, was “somewhat
important,” in only three sites (i.e., Knoxville, Albuquerque, and St. Petersburg) did (at least) 60
percent of the respondents believe that getting promoted to a higher rank was important (i.e.,
combining “somewhat” and “very” responses). Of interest, is that in the five other sites, there
was a greater percentage of officers that responded that such a promotion was “very
unimportant” compared to those that believed it was “very important.”
Whereas respondents were somewhat split in the importance of moving upward in their
occupation, we found much more favorable attitudes toward moving to a specialized unit. In
every agency, but Portland, the majority (and in many cases two-thirds and higher) believed it
was important, in varying degrees, to be promoted laterally. While Knoxville officers placed a
higher importance, of all sites, on vertical promotions (i.e., a higher percentage of respondents
reported it was important overall as well as “very”), St. Petersburg lead the way with respect to
specialized units. In fact, while every agency reported higher percentages of “very important”
responses in moving across the organization, compared to moving up, no agency came close to
the 40.6 percent found among St. Petersburg respondents. Based on these results, it is evident

87

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

that officers are aspiring, at least at this point in time, to moving to a specialized unit over
moving up in rank.
Finally, as another indicator of their future aspirations, we asked respondents to indicate
which rank they expected to hold when they retired. Given the varying vertical differentiation
found in our agencies, producing varying ranks structures, some of the classifications listed were
relevant for some departments, but not for others. In cases where a given rank did not exist in
that particular agency, an asterisk was placed in the cell.36 We found that roughly one-fifth, in
Fort Wayne, to two-fifths, in Columbus, believed that they would be retiring at the patrol rank.
At the other end of the spectrum, few respondents (ranging from 1.6% in Columbus to 7.4% in
Knoxville) believed that they would retire at the assistant/deputy chief or chief levels. With the
exception of Portland, who tended to split responses because of their unique “detective” response
option, and Knoxville, the majority of respondents (across all agencies) believed that they would
retire at a middle-level management rank (ranging from 60.9% in Colorado Springs to 50.4% in
Columbus).37

36

Portland, unlike the rest of our agencies, regarded “criminalists” and “detectives” as advancements in rank. As
such, their officers were the only ones for which we reported their responses to these positions. Each agency’s
survey was tailored, for this question, to reflect the various ranks that existed in that departments. There were some
officers, from other agencies, that responded “other” for this question, and wrote in (as instructed) that they expected
to retire at the rank of detective/investigator. Since these were not officially designated ranks for their agency, and
because all officers were not afforded the same opportunity to express the response of detective/investigator, we left
these responses among the “other” category (found in the last row of Table 5-16). Among other examples of listed
“other” ranks when retired, across sites, were: sheriff, inspector, community coordinator, arrest warrant officer,
firefighter, etc.
37

Because Charlotte-Mecklenburg was the only agency without a lieutenant rank, and because organizationally their
captains had many of the same duties and responsibilities as lieutenants in other agencies, we added captains and
sergeants together to draw these conclusions regarding middle-level management.

88

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 5-16. Officer Aspirations
CSPD PPB

APD

FWPD

SPPD

KPD

CMPD

CPD

43
(21.6)
55
(27.6)
71
(35.7)
30
(15.1)

68
(27.5)
71
(28.7)
91
(36.9)
17
(6.9)

44
(13.6)
81
(25.0)
142
(43.8)
57
(17.6)

23
(12.1)
72
(37.9)
76
(40.0)
19
(10.0)

34
(15.2)
54
(24.1)
93
(41.5)
43
(19.2)

20
(12.8)
32
(20.5)
63
(40.4)
41
(26.3)

86
(18.6)
130
(28.1)
179
(38.8)
67
(14.5)

123
(23.7)
176
(34.0)
183
(35.4)
36
(6.9)

24
(12.1)
39
(19.6)
88
(44.2)
48
(24.1)

44
(17.8)
84
(34.0)
67
(27.1)
52
(21.1)

30
(9.3)
61
(18.8)
145
(44.8)
88
(27.2)

24
(12.6)
69
(36.4)
66
(34.7)
31
(16.3)

24
(10.7)
26
(11.6)
83
(37.1)
91
(40.6)

16
(10.3)
25
(16.1)
68
(43.9)
46
(29.7)

58
(12.6)
85
(18.5)
203
(44.1)
114
(24.8)

51
(9.8)
110
(21.2)
244
(47.0)
114
(22.0)

68
(28.3)
3
(1.3)
49
(20.4)
56
(23.3)
32
(13.3)
*

83
(26.0)
*

38
(20.2)
*

73
(34.3)
*

57
(38.5)
*

146
(32.2)
*

194
(39.2)
*

*

*

*

*

*

*

103
(32.3)
88
(27.6)
20
(6.3)
*

72
(33.8)
41
(19.2)
*

138
(27.8)
112
(22.6)
*

*

22
(14.9)
19
(12.8)
18
(12.2)
*
(4.2)
*

163
(36.0)
*

*

86
(45.8)
28
(14.9)
23
(12.2)
*
(4.2)
*

*

*

*

*

Higher Rank
Very Unimportant
Somewhat Unimp.
Somewhat Imp.
Very Important

Specialized Unit
Very Unimportant
Somewhat Unimp.
Somewhat Imp.
Very Important

Expected Rank
When Retire
Patrol Officer
Criminalist
Detective
Sergeant
Lieutenant
Captain
Major
Precinct Commander

58
(29.1)
*
*
84
(42.3)
37
(18.6)
*
*

*
(3.8)
Division Commander 11
(5.5)
Assist./Dep. Chief
1
(0.5)
Chief
3
(1.5)
Other
5
(2.5)

*
9
*
1
(0.4)
10
(4.2)
12
(5.0)

3
(0.9)
6
(1.9)
16
(5.0)

4
(2.1)
5
(2.7)
4
(2.1)

89

9

1
(0.5)
8
(3.8)
9
(4.2)

4
(2.7)
7
(4.7)
21
(14.2)

75
(16.6)
19
*
*
6
(1.3)
8
(1.8)
36
(7.9)

*
*
21
(4.2)
0
(0)
8
(1.6)
23
(4.6)

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The purpose of the final section of the patrol officer survey was to capture a variety of
background and demographic factors (some of which were gleaned from other official
organizational data), as well as future aspirations of respondents. Across the various factors, we
found both internal and external variation by site. At the same time, some patterns were
revealed.
The majority of all respondents were white, married (with the exception of St.
Petersburg), males, with no military experience, no relatives in policing, and had at least some
college experience. Among the respondents that did have military and/or college backgrounds,
the more powerful experiences (in terms of preparing to be a police officer) were noted for the
military. The average age of respondents was roughly around 35 years old, and the oldest
officers were found among Colorado Springs and Portland respondents. Experience ranged from
less than a year to 38 years, while Albuquerque, Knoxville, and St. Petersburg had the least
experienced (organizationally and overall) respondents.
In terms of overall police aspirations, officers (across all sites) expressed more of an
interest in moving to a specialized unit over a promotion in rank. Finally, the modal response
indicated that the majority of all respondents believed that they would ultimately retire at some
type of middle management position.

90

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 6
Force Report Data

In addition to surveying officers about their agency’s use of force policy, a key part of
the second phase of the project involved the collection and analysis of three pieces of official
data: use of force reports, citizen complaints, and lawsuits. In this chapter we examine force
reports to investigate which policy types are associated with varying degrees of force, as well as
injuries to suspects and officers.

Methodology
At each of the sites we collected 24 months of force report data. The time frame varied
to some degree from site to site to ensure we had a consistent time period with no policy changes
either in substance (i.e., the content or type of policy changed, such as altering tactical
placement) or reporting (i.e., the force report form changed or the threshold for reportable force
was altered)(see footnote 26 for exact timeframe). Some of the research sites coded nearly all
data electronically (i.e., Portland, Charlotte-Mecklenburg, Columbus), others only maintained
paper copies (i.e., Colorado Springs, Fort Wayne, Knoxville), and yet others had some sort of
combination between electronic data and paper copies (Albuquerque, St. Petersburg). Cities that
captured use of force behavior solely on paper required us to code the data into electronic
format. Even in cities where some (or most) of the data were already electronically coded, we
often had to go back to the hard copy originals to adequately code all the variables we were
interested in (i.e., parts of the force forms were electronically coded by the agencies, but had to
be supplemented with information from hard copies by project staff). For example, in

91

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Albuquerque, the type of force used was captured in such a way that we were unable to compare
force types with that used in other cities simply due to the way it was coded in their dataset.
Hence, we had to make copies of paper reports and then match them to the electronic cases to
clarify the type of force used and code into the dataset.
While we were interested in a number of relevant situational and contextual variables
from the force report data, the key variables we wanted to capture involved the types of force
used in response to the types of resistance suspects presented to officers. Importantly, we
wanted to ensure that the types of force being coded from each city were comparable. As such,
we created “common measures” of resistance and force across each of the cities. For citizen
resistance, the following categories were classified. Compliance involved suspect behavior that
responded to officer direction without resistance. Passive resistance was defined as suspect
behaviors that were unresponsive to police verbal communication or direction. Behaviors in this
category must not have been verbalized or involve physical movements in a defensive or
aggressive mode, but rather demonstrated through inactivity (e.g., ignored or disregarded police
attempts at verbal communication or control, went limp, failed to physically respond or move).
Verbal resistance included a suspect verbally rejecting police verbal communication or direction.
Suspect behaviors in this category must have been verbal in nature and not physical (e.g., telling
the officer he or she will not comply with police direction, to leave alone, or not to bother him or
her). Defensive resistance was defined as suspect attempts to evade police attempts at control.
Suspect behaviors in this category must have involved some type of physical behavior in a
defensive mode (e.g., attempts to leave the scene, flee, hide from detection, pull away from
officer’s grasp). Active resistance included the suspect either attempting or actually attacking or

92

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

striking an officer. Suspect behaviors in this category must have involved some type of physical
behavior in an aggressive mode (e.g., lunging toward the police, striking police with hands, fists,
kicks or any instrument that may perceived as a weapon such as a knife, stick, frying pan).
Finally, deadly resistance was classified as attempts or actual attacks that could cause death.
Police use of force behavior was split into weaponless tactics and weapon tactics. For
weaponless tactics, the following categories were classified. First, we coded handcuffing when
it was noted in the dataset or on the use of force report forms. A firm grip involved an officer
grabbing/holding a suspect, which could also involve an escort whereby an officer physically
guides a suspect’s movement (e.g., placed a suspect into a patrol car). This type of force is
simply low level hands on force where the officer is touching or guiding a suspect, but not
struggling or being aggressive with the suspect. Pressure points (i.e., pain compliance
techniques) involved holds that cause pain to a specific body part (e.g., hammerlock, wristlock,
finger grip). Control maneuvers went a step beyond a firm grip or pain compliance technique
and involved an officer using hands on force to attempt to gain control of a suspect (e.g., hands
on struggling to get a suspect’s hands cuffed, an arm bar technique, bear hug, etc.). Takedowns
involved instances when suspects were thrown, pushed, or shoved to the ground, against a wall,
against a car or any other surface. Finally, empty hand strikes with the body included
hitting/striking a suspect with the hands, fists, feet, legs, or any other part of the body (e.g.,
slapping, punching, kicking). For weapon tactics, the following categories, all self-defined, were
classified as chemicals spray, baton, CED, impact munitions (e.g., beanbag), and firearm.38

38

We also coded officer pointing of the firearm when noted on the use of force report form.

93

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Analyses and Findings
Frequency of Force by Workload Measures
We begin by examining how often officers used force across the eight cities. In addition,
we examine how often force is used in relation to three workload measures: calls for service,
reported crimes, and arrests. As illustrated in Table 6-1 (bottom row), the range of force usage
over a two year period ranges from 738 in Colorado Springs to 5,787 in Columbus. Comparing
force usage by standardized calls for service we see there is substantial variation across the
cities. Officers in Colorado Springs, for instance, used force 1.4. times for every 1,000 calls for
service. Viewed alternatively, officers used force once for every 729 calls for service.
Contrasted with Portland (the city with the most frequent force usage in relation to calls for
service) officers used force 9.8 times for every 1,000 calls for service, or once for every 102 calls
for service. In other words, when using calls for service as a comparison point, officers in
Portland used force seven times more frequently than officers in Colorado Springs.
A good deal of variation is also present when comparing cities by the amount of reported
crimes. Here, Charlotte-Mecklenburg used force the least frequently. Officers used force once
for every 87 reported Part I index crimes filed. Conversely, Fort Wayne used force once for
every 12 Part I index crimes. Hence, officers in Fort Wayne used force over seven times more
frequently than officers in Charlotte-Mecklenburg.

94

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-1. Force Usage Comparison by Calls for Service, Reported Crimes, Arrests39
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

1.4

9.8

2.2

5.6

5.3

2.1

3.8

4.6

1 of 729

1 of 102

1 of 447

1 of 177

1 of 189

1 of 481

1 of 264

1 of 608

19.0

60.5

20.6

86.1

48.4

34.7

49.0

11.5

1 of 52

1 of 16

1 of 48

1 of 12

1 of 21

1 of 29

1 of 20

1 of 87

83.3

305.6

231.1

403.6

302.2

154.7

427.8

93.7

Force per arrest

1 of 12.0

1 of 3.2

1 of 4.4

1 of 2.5

1 of 3.3

1 of 6.5

1 of 2.3

1 of 9.0

Total Number
Force Reports

738

4374

1458

1997

1897

926

5787

1332

Force/Service Calls
(1,000)
Force per call
Force/Part I Crimes
(1,000)
Force per crime
Force/Part I Arrests
(1,000)

39

Calls for service, crime, and arrest totals compiled using a variety of sources including annual reports and raw data provided by the police departments. Calls
for service totals (CSPD=538,280, PPB=446,869, APD=652,366, FWPD=354,175, SPPD=357,709, KPD=446,161, CPD=1,528,280, CMPD=810,423); Part 1
crime totals (CSPD=38,793, PPB=72,278, APD=70,683, FWPD=23,196, SPPD=39,181, KPD=26,652, CPD=117,982, CMPD=115,663); arrest totals
(CSPD=8,857, PPB=14,309, APD=6,308, FWPD=4,948, SPPD=6,277, KPD=5,985, CPD=13,528, CMPD=12,060).

95

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Finally, substantial variation was also unveiled when comparing force usage by arrests.
Similar to calls for service, Colorado Springs used force the least frequently in terms of arrests
(one force report filed per every 12 arrests made). On the other hand, Columbus officers used
force most frequently (one force report filed per every 2.3 arrests made). Thus, officers in
Columbus used force over five times more frequently than officers in Colorado Springs.

Force and Resistance Types
Next, we turn to a breakdown of physical force types (weaponless tactics) as depicted in
Table 6-2. Figures are computed by the type of force used in relation to the total number of
force reports filed per city. Fort Wayne used the most physical force at 91.1 percent. This is not
overly surprising given that this is the only city where everyday patrol officers do not carry a
CED. At the opposite end of the continuum, just over half (53.9%) of the force reports filed by
Colorado Springs officers involved a weaponless tactic.
In terms of a firm grip, some cities (i.e., St. Petersburg and Charlotte-Mecklenburg)
report using a substantially higher number than others (i.e., Portland and Albuquerque) who
report very few such uses of force. It appears much of this variation is a function more in terms
of reporting procedures than actual differences in force use at this level. According to
departmental officials in St. Petersburg, for instance, officers routinely report this level of hands
on force when using it. In Portland, however, officials report that despite using this form of
force regularly, officers often do not report it when used in conjunction with other more serious
forms of force.

96

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-2. Frequency of Types of Physical Force (Weaponless) Used
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Any
Physical

398
(53.9)

3430
(78.4)

967
(66.3)

1820
(91.1)

1415
(74.6)

804
(86.8)

4532
(78.3)

1156
(86.8)

Firm
Grip

39
(5.3)

3
(0.1)

17
(1.2)

167
(8.4)

787
(41.5)

209
(22.6)

640
(11.1)

529
(39.7)

Pressure
Point

41
(5.6)

367
(8.4)

15
(1.0)

105
(5.3)

108
(5.7)

101
(10.9)

509
(8.8)

133
(10.0)

Control
Maneuver

256
(34.7)

2181
(49.9)

493
(33.8)

1092
(54.7)

402
(21.2)

579
(62.5)

2289
(39.6)

1030
(77.3)

Take
Down

172
(23.3)

1796
(41.1)

670
(46.0)

910
(45.6)

786
(41.4)

440
(47.5)

2261
(39.1)

732
(55.0)

Strike

153
(20.7)

541
(12.4)

242
(16.6)

640
(32.0)

133
(7.0)

183
(19.8)

398
(6.9)

211
(15.8)

Other

0
(0.0)

233
(5.3)

17
(1.2)

0
(0.0)

1
(0.1)

0
(0.0)

18
(0.3)

0
(0.0)

Total

738

4374

1458

1997

1897

926

5787

1332

97

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

With respect to other soft hand tactics, Knoxville reports using the most pressure points (10.9%),
while Charlotte-Mecklenburg reports using the most control maneuvers (77.3%) and take down
(55.0%) techniques. Finally, Fort Wayne officers used hard hand striking tactics in nearly one of
every three force reports (32.0%). The city with the next most frequent hard hand strikes is
Colorado Springs with 20.7 percent. Both St. Petersburg and Columbus reported using such
force relatively infrequently (7.0% and 6.9%, respectively).
In Table 6-3 we examine a breakdown of weapon based force types. Once again, figures
are computed by the type of force used in relation to the total number of force reports filed per
city. Colorado Springs used the most weapon force at 62.7 percent, while Fort Wayne used
some form of a weapon in just over 25 percent. When looking at this table as a whole it is
evident that a bulk of the weapon force used are chemical sprays and CEDs. Baton and impact
munitions are relatively infrequent. For example, Albuquerque used the baton most often yet
such force was only used in 2.7 percent of the cases. Looking at chemical sprays, Albuquerque
and Fort Wayne rely on such force in 24.8 percent and 22.1 percent of their force encounters
respectively, but Charlotte-Mecklenburg, Knoxville, and Portland all report roughly four to five
percent.
Perhaps the type of force that stands out most in Table 6-3 is the CED. In particular,
Colorado Springs reported using a CED in half of all force encounters (49.5%). The city with
the next highest frequency of use is Portland at 23.4 percent, or less than half what Colorado
Springs reports. Understandably, Fort Wayne reports very few deployments of a CED (only a
hand full of officers are CED equipped).

98

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-3. Frequency of Types of Weapons Used
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Any
Weapon

463
(62.7)

1553
(35.5)

720
(49.4)

511
(25.6)

613
(32.3)

274
(29.6)

1678
(29.0)

398
(29.9)

Chemical
Spray

91
(12.3)

248
(5.7)

361
(24.8)

442
(22.1)

202
(10.6)

44
(4.8)

1049
(18.1)

56
(4.2)

Baton

3
(0.4)

37
(0.8)

39
(2.7)

45
(2.3)

3
(0.2)

21
(2.3)

34
(0.6)

15
(1.1)

CED

365
(49.5)

1023
(23.4)

305
(20.9)

9
(0.5)

406
(21.4)

193
(20.8)

600
(10.4)

252
(18.9)

Impact
Munitions

0
(0.0)

122
(2.8)

13
(0.9)

12
(0.6)

0
(0.0)

1
(0.1)

0
(0.0)

6
(0.5)

Point

7
(0.9)

351
(8.0)

17
(1.2)

0
(0.0)

32
(1.7)

16
(1.7)

18
(0.3)

57
(4.3)

Deadly

6
(0.8)

18
(0.4)

0
(0.0)

1
(0.1)

6
(0.3)

0
(0.0)

0
(0.0)

15
(1.1)

Other

10
(1.4)

175
(4.0)

48
(3.3)

15
(0.8)

17
(0.9)

30
(3.2)

38
(0.7)

71
(5.3)

Total

738

4374

1458

1997

1897

926

5787

1332

99

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Finally, Portland reported a substantially higher use of both firearm pointing and other forms of
force not clearly specified in the force reports. For example, Portland officers reported pointing
their firearms in 8.0 percent of the cases. The next highest city was Charlotte-Mecklenburg at
4.3 percent.
Table 6-4 provides a look at the highest use of force applied. For ease of presentation,
the categories of force have been collapsed into the following categories: soft hands (i.e., firm
grip, pressure points, control maneuver, take downs), hard hands (i.e., empty hand strikes) and
chemical spray, and CED/impact (i.e., CED, baton, munitions). Note that ranking force usage
according to the highest level is a difficult proposition. As the national agency survey results
indicated in Chapter 3, as well as the eight selected cities in Phase II, there is no commonly
accepted ranking of force. Nonetheless, based on the national agency survey results, the most
common ranking used conforms with the ranking used in Table 6-4, and hence the reason for
presenting in this order. As shown in this table, Columbus is most likely to use the lowest form
of force (soft hands, 65.1%); Fort Wayne most likely to use the intermediate level (hard
hands/chemical spray, 47.1%), and Colorado Springs most likely to use the highest level
(CED/impact, 51.3%). There were very few instances of deadly force in any of the cities.
Conversely, Colorado Springs used the least soft hands (22.0%), Portland the least hard
hands/chemical sprays (14.5%), and Fort Wayne the least CED/impact (4.1%).

100

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-4. Officer Force Types (Highest)
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Soft
Hands

162
(22.0)

2353
(54.5)

533
(36.7)

974
(48.8)

970
(57.5)

498
(54.6)

3707
(65.1)

752
(56.5)

Hard Hands/
Chemical Spray

191
(25.9)

625
(14.5)

524
(36.1)

941
(47.1)

293
(17.4)

177
(19.4)

1320
(22.8)

227
(17.1)

CED/
Impact

378
(51.3)

1318
(30.6)

394
(27.2)

81
(4.1)

418
(22.0)

237
(25.6)

665
(11.7)

336
(25.3)

Deadly

6
(0.8)

18
(0.4)

0
(0.0)

1
(0.1)

6
(0.3)

0
(0.0)

0
(0.0)

15
(1.1)

Total

737

4314

1451

1997

1687

912

5692

1330

101

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-5 takes a look at citizen resistance.40 Nearly all the cities report high levels of
citizen resistance. With the exception of Portland, all the cities are in the high ninety percent
range or report all cases where citizens resisted in some manner. When looking at the types of
resistance reported, there is a vast degree of variation in failure to comply, with less variation,
although still quite substantial, in relation to defensive and aggressive physical. In terms of
failure to comply (i.e., passive and verbal resistance), Columbus shows that nearly every force
case involved this type of resistance, while Albuquerque and Fort Wayne are in the 30 percent
range. For physically defensive resistance, Colorado Springs, St. Petersburg, Knoxville, and
Charlotte-Mecklenburg are all in the 70 to 80 percent range, while Portland and Albuquerque are
in the 40 percent range. Overall, variation is less evident from a relative perspective for
aggressive physical resistance. With the exception of St. Petersburg, the range of this form of
resistance across the remaining seven cities is 31 to 51 percent. Hence, Knoxville (51.3%)
reports more than two and a half times the amount of aggressive physical resistance compared to
St. Petersburg (20.0%).

40

While Portland captured suspect resistance on their use of force report form, they did so in such a way that we
were unable to create and match up the types of resistance reported with our common measures of resistance. As a
result, we needed to secure hard copies of their incident/investigation reports where the narrative portion of the force
incident was described. However, obtaining these reports, due to sheer volume and departmental workload
limitations, proved difficult. Nonetheless, we were only able to get one year’s worth of reports. From these, we
were then able to code suspect resistance in accordance with our common measures definitions.

102

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-5. Suspect Resistance Types (Any)
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Any
Resistance

738
(100.0)

1868
(89.8)

1458
(100.0)

1997
(100.0)

1850
(97.5)

919
(99.2)

5707
(96.6)

1326
(99.5)

Fail to
Comply

305
(41.3)

1595
(76.7)

507
(34.8)

749
(37.5)

881
(46.4)

392
(42.3)

5630
(97.3)

716
(53.8)

Defensive
Physical

568
(77.0)

953
(45.5)

723
(49.6)

1303
(65.2)

1381
(72.8)

734
(79.3)

3870
(66.9)

1112
(83.5)

Aggressive
Physical

233
(31.6)

680
(32.7)

707
(48.5)

619
(31.0)

380
(20.0)

475
(51.3)

1965
(34.0)

561
(42.1)

Deadly

8
(1.1)

0
(0.0)

35
(2.4)

28
(1.4)

24
(1.3)

36
(3.9)

103
(1.8)

23
(1.7)

Total

738

2080

1458

1997

1897

926

5787

1332

* 1-year

103

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Similar to Table 6-4 for the highest level of physical force used, Table 6-6 looks at the
highest level of resistance posed across the eight cities. As shown in this table, Columbus
reported the most failure to comply cases as the highest form of suspect resistance (25.7%); Fort
Wayne was most likely to report defensive physical resistance (64.0%), and Knoxville most
likely to report aggressive physical resistance (48.8%). There were much fewer instances of
deadly resistance in any of the cities, with Knoxville reporting the most (3.9%).
Given the somewhat difficult manner in which to interpret how officers are using force
compared to resistance based on the series of tables thus far, Tables 6-7a and 6-7b offer a more
consolidated way to interpret the highest levels of force and resistance across the eight cities.
Using the ranking of force offered in Table 6-4 (0=soft, 1=hard/chemical, 2=CED/impact,
4=deadly), Table 6-7a shows that Colorado Springs had the highest mean level of force (1.31),
while Columbus had the least (0.47). Similarly, using the ranking of resistance offered in Table
6-6 (0=none, 1=fail to comply, 2=defensive, 3=aggressive, 4=deadly), Table 6-7b shows that
Albuquerque had the highest mean level of resistance (2.53), while Portland had the least (1.92).
When looking at Tables 6-7a and 6.7b in tandem, one can see that for some of the cities the
ranking of resistance and force are closely similar. For example, Albuquerque officers used the
second most force but faced the most resistant suspects. Moreover, Charlotte-Mecklenburg
ranks fourth on force usage and third in terms of resistance. However, some of the cities show
more of a disconnect between the level of force used in relation to the amount of resistance
faced, at least in the aggregate as depicted here. For instance, Colorado Springs officers used the
highest amount of force, but only encountered the fifth most (i.e., highest level of) resistance.

104

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-6. Suspect Resistance Types (Highest)
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Fail to
Comply

58
(7.9)

414
(19.9)

8
(0.5)

75
(3.8)

336
(17.7)

32
(3.5)

1489
(25.7)

40
(3.0)

Defensive
Physical

439
(59.5)

774
(37.2)

708
(48.6)

1279
(64.0)

1117
(58.9)

399
(43.1)

2213
(38.2)

711
(53.4)

Aggressive
Physical

233
(31.6)

680
(32.7)

707
(48.5)

615
(30.8)

373
(19.7)

452
(48.8)

1902
(32.9)

552
(41.4)

Deadly

8
(1.1)

0
(0.0)

35
(2.4)

28
(1.4)

24
(1.3)

36
(3.9)

103
(1.8)

23
(1.7)

Total

738

2080

1458

1997

1897

926

5787

1332

* 1-year

105

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-7a. Aggregate Mean Levels of Force (Highest)

Mean
(Rank)

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

1.31
(1)

0.77
(3)

0.90
(2)

0.55
(7)

0.68
(6)

0.70
(5)

0.47
(8)

0.71
(4)

Table 6-7b. Aggregate Mean Levels of Resistance (Highest)

An
(Rank)

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

2.26
(5)

1.92
(8)

2.53
(1)

2.30
(4)

2.00
(7)

2.52
(2)

2.08
(6)

2.42
(3)

106

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Portland shows a similar degree of disconnect (i.e., using substantially more force than the
resistance faced). Several cities, however, particularly Fort Wayne and Knoxville, exhibit a
reverse pattern. For instance, Knoxville officers report the second highest degree of resistance
from suspects yet only used the fifth highest amount of force. While these comparisons do not
offer a full picture concerning the individual micro-level dynamics of resistance/force, they do
provide a big picture comparison in terms of where cities stand in the aggregate.

Bivariate Analyses
Next we consider how different departments instruct officers (via force policy) on the
type of force that should be used in relation to the type of subject resistance faced so we can dig
a little deeper to examine how officers are using various forms of force within their individual
policy context. Table 6-8 shows the six departments in our study that link resistance to force.
The other two cities (Fort Wayne and Columbus) do not link resistance to force via their policy,
but rather simply lay out the force continua with no explicit connection to specific types of
resistance. Interestingly, Colorado Springs, who employs a wheel force continuum design for
the stated purpose that they do not want officers to think “linearly,” still connects resistance to
force within the text of their policy.

107

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-8. Allowable Force by Resistance Type41
Col. Sp.

Port.

Char.

St. Pete

Knox.

Albq.

Verbal

Soft

Verbal

Verbal

Soft

Soft

Soft
Chem
Hard
CED

Passive

Soft

Soft

Soft

Soft

Soft

Soft
Chem
Hard

Physical
Defensive

Chem
CED

Chem
CED

Chem
Hard

Chem
Hard
Impact

Chem
Hard
CED
Impact

Soft
Chem
Hard
CED

Physical
Aggressive

Hard
Impact

Hard
Impact

CED
Impact

CED

Chem
Hard
CED
Impact

Impact

Resistance

41

Of course, lesser forms of force are always permitted as one moves up for the force continuum. For instance, in Colorado Springs chemical spray and a CED
are permissible forms of force when officers are faced with physically defensive resistant suspects, but officers may also use a lesser form of force as well (in this
case, soft hand force). Similarly, officers facing physically aggressive suspects may use any form of force (soft, chemical spray, CED, hard, impact) short of
deadly force.

108

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

We begin by examining how officers in each of the cities responded to those suspects
whose highest level of resistance throughout the encounter was failure to comply (i.e., passive
and verbal resistance). As illustrated in Table 6-9, a majority of these encounters result in
officers using some type of soft hand tactic (i.e., firm grip, control maneuver, pressure points,
take downs). Of the cities that linked force to resistance, only Albuquerque allowed officers to
use anything more coercive than soft hand tactics on suspects who verbally or passively
resisted.42 In other words, Albuquerque allowed officers to use hard hand tactics, chemical
spray, and/or the CED on those who non-physically failed to comply. Despite this, officers in
Albuquerque still relied quite a bit on soft hand tactics.
Table 6-10 also depicts those encounters when officers were faced with suspects who
failed to comply, but in this instance responded with hard hand tactics (i.e., striking the suspects
with hand, fist, feet). Fort Wayne officers were most likely to use hard hand tactics (22.7%) in
these situations, but their policy does not specify that officers are prohibited from such force
(recall their policy does not link resistance and force). Colorado Springs officers used hard hand
strikes in roughly one of every five (19.0%) such encounters, despite their policy restricting such
usage (Table 6-8). Conversely, Albuquerque, whose policy allows such force, only reported
facing eight encounters of this type, but in none of them did officers report using a hard hand
strike.

42

When we refer to agency policies “allowing” or “permitting” various types of force given various types of
resistance, we are assuming that such resistance stands alone (i.e., occurred once during the encounter). However,
we know from prior work (Terrill, 2005) that suspects sometimes resist multiple times with the same resistance type
within the same encounter, which would affect whether an officer is operating within or outside policy parameters.
For example, an officer may use a hard hand tactic on a suspect whose highest level of resistance never rises above
failure to comply, which would appear to be out of policy (e.g., in Colorado Springs and Portland). However, it is
possible that a given suspect will repeatedly fail to comply prompting an officer to work his/her way up the
continuum to a hard hand tactic, which would in effect be within policy in this scenario. As a result, one must use
caution with interpretation of the findings we present here.

109

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-9. Suspects Highest Level of Resistance was Fail to Comply
Did Officer Use Soft Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

25
(43.1)

270
(65.2)

5
(62.5)

57
(76.0)

222
(66.1)

24
(75.0)

880
(59.1)

34
(85.0)

No

33
(56.9)

144
(34.8)

3
(37.5)

18
(24.0)

114
(33.9)

8
(25.0)

609
(40.9)

6
(15.0)

58

414

8

336

32

1489

1489

40

Total
* 1-year

Table 6-10. Suspects Highest Level of Resistance was Fail to Comply
Did Officer Use Hard Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

11
(19.0)

29
(7.0)

0
(0.0)

17
(22.7)

13
(3.9)

2
(6.2)

24
(1.6)

1
(2.5)

No

47
(81.0)

385
(93.0)

8
(100)

58
(77.3)

323
(96.1)

30
(93.8)

1465
(98.4)

39
(97.5)

58

414

8

75

336

32

1489

50

Total
* 1-year

110

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-11 looks at failure to comply with chemical spray usage. Once again, of the
eight cases in Albuquerque, the one city that permits this type of force on this type of resistant
suspect, none resulted. Columbus had 31.4 percent of its cases result in chemical spray use
followed by St. Petersburg at 23.5 percent. For CED use (Table 6-12), Albuquerque officers
used such force in three of its eight cases (37.5%), while Colorado Springs officers used a CED
on 26 of 58 passive or verbally resistant suspects (44.8%). Table (6-13) depicts cases where
suspects failed to comply as their highest form of resistance and shows few instances where
officers used a baton. Knoxville was the highest at 9.4 percent, but this consisted of just 3 of 32
cases.
Overall, when looking at instances where a suspect’s highest level of resistance was
failure to comply (Tables 6-9 to 6-13), it appears a department’s policy has little effect. For
instance, while the number of cases are low (N=8), we see that Albuquerque officers were not
necessarily using higher forms of force (i.e., hard hands, chemical spray, CED) on such suspects
even though their policy allowed it. Conversely, we see officers in other cities using forms of
force that were restricted from being used on failure to comply suspects. For example, Colorado
Springs officers used a CED on passive or verbally resistant suspects in 44.8 of the cases, while
Knoxville officers used such force in 21.9 of the cases.

111

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-11. Suspects Highest Level of Resistance was Fail to Comply
Did Officer Use Chemical Spray?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

7
(12.1)

19
(4.6)

0
(0.0)

13
(17.3)

79
(23.5)

0
(0.0)

467
(31.4)

2
(5.0)

No

51
(87.9)

395
(95.4)

8
(100)

62
(82.7)

257
(76.5)

32
(100)

1022
(68.6)

38
(95.0)

58

414

8

75

336

32

1489

40

Total
* 1-year

Table 6-12. Suspects Highest Level of Resistance was Fail to Comply
Did Officer Use CED?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

26
(44.8)

80
(19.3)

3
(37.5)

0
(0.0)

23
(6.8)

7
(21.9)

110
(7.4)

5
(12.5)

No

32
(55.2)

334
(80.7)

5
(62.5)

75
(100)

313
(93.2)

25
(78.1)

1379
(92.6)

35
(87.5)

58

414

8

75

336

32

1489

40

Total
* 1-year

112

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-13. Suspects Highest Level of Resistance was Fail to Comply
Did Officer Use Baton?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

0
(0.0)

1
(0.2)

0
(0.0)

1
(1.3)

0
(0.0)

3
(9.4)

2
(0.1)

0
(0.0)

No

58
(100.0)

413
(99.8)

8
(100.0)

74
(98.7)

336
(100.0)

29
(90.6)

1487
(99.9)

40
(100.0)

58

414

8

75

336

32

1489

40

Total
* 1-year

113

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The next set of tables (6-14 to 6-18) take a similar approach to the last set, but in these
cases the form of resistance is physical defensive (i.e., physical behaviors to evade police
attempts at control such as bracing, pulling away, and/or fleeing from an officer). To consider
the type of force used in relation to the type of resistance faced, it is again important to place the
findings into the context of Table 6-8 showing what each city’s policy permits. For instance,
soft hands are permitted in all the cities for this type of resistance, but other forms of force are
allowed (or not) depending on the city. For example, Colorado Springs and Portland allow
officers to use chemical spray and a CED, but not hard hand strikes nor impact methods (e.g.,
baton blows or munitions). Other cities mix and match different variations for this level of
resistance.
Similar to failure to comply cases previously, a majority of physically defensive
encounters result in officers using some type of soft hand tactics (Table 6-14). Again, Colorado
Springs is the only city in which officers responded to such resistance with soft hand tactics in
less than 50 percent of the cases. When the results are examined for hard hand tactics (Table 615) it is difficult to discern a pattern based on policy differences. For example, in Portland, a
city that does not permit hard hand strikes on defensively resistant suspects, the frequency of use
was 13.0 percent. However, in three of the cities that permit hard hand strikes (Albuquerque,
Knoxville, and Charlotte) the percentages are very similar at 14.7 percent, 15.3 percent, and 14.9
percent, respectively. Colorado Springs, which similar to Portland, restricts hard hand strikes on
defensively resistant suspects, used such force in 22.1 percent of its encounters. The highest
frequency city was Fort Wayne (31.7%), which does not link force to resistance.

114

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-14. Suspects Highest Level of Resistance was Physical Defensive
Did Officer Use Soft Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

199
(45.3)

607
(78.4)

466
(65.8)

1090
(85.2)

823
(73.7)

327
(82.0)

1848
(83.5)

621
(87.3)

No

240
(54.7)

167
(21.6)

242
(34.2)

189
(14.8)

294
(26.3)

72
(18.0)

365
(16.5)

90
(12.7)

439

774

708

1279

1117

339

2213

711

Total
* 1-year

Table 6-15. Suspects Highest Level of Resistance was Physical Defensive
Did Officer Use Hard Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

97
(22.1)

101
(13.0)

104
(14.7)

405
(31.7)

70
(6.3)

61
(15.3)

77
(3.5)

106
(14.9)

No

342
(77.9)

673
(87.0)

604
(85.3)

874
(68.3)

1047
(93.7)

338
(84.7)

2136
(96.5)

605
(85.1)

439

774

708

1279

1117

399

2213

711

Total
* 1-year

115

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

With the exception of Albuquerque (18.4%) and Fort Wayne (18.1%), the use of
chemical spray was fairly infrequent on physically defensive suspects as shown in Table 6-16.
Alternatively, CED use (Table 6-17) was more frequent in a number of cities. Colorado Springs
used a CED far more frequently in these circumstances (50.1%) than any of the other cities.
Interestingly, the two cities that generally restrict the CED (St. Petersburg and CharlotteMecklenburg) actually diverge in terms of how often a CED is used in these types of resistant
cases. St. Petersburg, who instructs officers to use a CED on physically aggressive resistant
suspects (but not physical defensive), relied on such force 23.9 percent of the time. Conversely,
Charlotte-Mecklenburg, who has a similar, although slightly less restrictive, policy used the
CED on just 10.5 percent of the cases. Finally, baton usage, like that on failure to resist
suspects, was rare on defensive resistant suspects as shown in Table 6-18. Fort Wayne was the
highest at just 2.0 percent. Six of the cities had less than 10 such cases of baton use across the
two year period.

116

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-16. Suspects Highest Level of Resistance was Physical Defensive
Did Officer Use Chemical Spray?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

44
(10.0)

31
(4.0)

130
(18.4)

232
(18.1)

82
(7.3)

10
(2.5)

202
(9.1)

18
(2.5)

No

395
(90.0)

743
(96.0)

578
(81.6)

1047
(81.9)

1035
(92.7)

389
(97.5)

2011
(90.9)

693
(97.5)

439

774

708

1279

1117

339

2213

711

Total
* 1-year

Table 6-17. Suspects Highest Level of Resistance was Physical Defensive
Did Officer Use CED?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

220
(50.1)

172
(22.2)

132
(18.6)

5
(0.4)

267
(23.9)

69
(17.3)

217
(9.8)

75
(10.5)

No

219
(49.9)

602
(77.8)

576
(81.4)

1274
(99.6)

850
(76.1)

330
(82.7)

1996
(90.2)

636
(89.5)

439

774

708

1279

1117

399

2213

711

Total
* 1-year

117

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-18. Suspects Highest Level of Resistance was Physical Defensive
Did Officer Use Baton?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

1
(0.2)

1
(0.1)

10
(1.4)

26
(2.0)

2
(0.2)

6
(1.5)

8
(0.4)

3
(0.4)

No

438
(99.8)

773
(99.9)

698
(98.6)

1253
(98.0)

1115
(99.8)

393
(98.5)

2205
(99.6)

708
(99.6)

439

774

708

1279

1117

399

2213

711

Total
* 1-year

118

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Next, we examine how officers in each of the cities responded to aggressively resistant
suspects (i.e., hostile or overt physical aggression towards and officer such as attempting or
actually striking an officer). Beginning with Table 6-19, we see that in all the cities except
Colorado Springs, a majority of the encounters involved soft hand tactics when dealing with
aggressive suspects. Knoxville had the highest amount at 87.4 percent while Colorado Springs
had the lowest at 39.9 percent. A substantial degree of variation is found when examining hard
hand tactics and aggressive resistance as displayed in Table 6-20. Here, Fort Wayne officers
resorted to hard hand tactics in 35.3 percent of their encounters while Columbus officers used
such force in just 13.9 percent of their encounters.
Moving to the use of weapons, Table 6-21 looks at chemical spray and aggressive
resistance. Albuquerque and Fort Wayne used the most chemical spray in these types of cases
(32.1% and 30.9%, respectively), while Charlotte-Mecklenburg and Knoxville used the least
(6.3% and 7.1 percent, respectively). With respect to CED usage, Table 6-22 shows Colorado
Springs officers were much more likely to resort to a CED when dealing with aggressively
resistant suspects compared to the other cities. Specifically, Colorado Springs officers used
CED force in 51.1 percent of these types of cases. The next closest city was Portland at 33.2
percent. Finally, Table 6-23 examines baton usage and aggressively resistant suspects. As
shown, baton usage on the whole was infrequent. Albuquerque (3.8%) and Fort Wayne (2.6%)
used the baton most, while St. Petersburg (0.3%) and Colorado Springs (0.9%) used it least.

119

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-19. Suspects Highest Level of Resistance was Physical Aggressive
Did Officer Use Soft Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

93
(39.9)

439
(64.6)

424
(60.0)

500
(81.3)

292
(78.3)

395
(87.4)

1510
(79.4)

479
(86.8)

No

140
(60.1)

241
(35.4)

283
(40.0)

115
(18.7)

81
(21.7)

57
(12.6)

392
(20.6)

73
(13.2)

233

680

707

615

373

452

1902

552

Total
* 1-year

Table 6-20. Suspects Highest Level of Resistance was Physical Aggressive
Did Officer Use Hard Hands?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

45
(19.3)

122
(17.9)

137
(19.4)

217
(35.3)

45
(12.1)

107
(23.7)

265
(13.9)

100
(18.1)

No

188
(80.7)

558
(82.1)

570
(80.6)

398
(64.7)

328
(87.9)

345
(76.3)

1637
(86.1)

452
(81.9)

233

680

707

615

373

452

1902

552

Total
* 1-year

120

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-21. Suspects Highest Level of Resistance was Physical Aggressive
Did Officer Use Chemical Spray?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

40
(17.2)

90
(13.2)

227
(32.1)

190
(30.9)

40
(10.7)

32
(7.1)

337
(17.7)

35
(6.3)

No

193
(82.8)

590
(86.8)

480
(67.9)

425
(69.1)

333
(89.3)

420
(92.9)

1565
(82.3)

517
(93.7)

233

680

707

615

373

452

1902

552

Total
* 1-year

Table 6-22. Suspects Highest Level of Resistance was Physical Aggressive
Did Officer Use CED?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

119
(51.1)

226
(33.2)

146
(20.7)

2
(0.3)

109
(29.2)

103
(22.8)

247
(13.0)

170
(30.8)

No

114
(48.9)

454
(66.8)

561
(79.3)

613
(99.7)

264
(70.8)

349
(77.2)

1655
(87.0)

382
(69.2)

233

680

707

615

373

452

1902

552

Total
* 1-year

121

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-23. Suspects Highest Level of Resistance was Physical Aggressive
Did Officer Use Baton?
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

2
(0.9)

14
(2.1)

27
(3.8)

16
(2.6)

1
(0.3)

11
(2.4)

20
(1.1)

11
(2.0)

No

231
(99.1)

666
(97.9)

680
(96.2)

680
(96.2)

372
(99.7)

441
(97.6)

1882
(98.9)

541
(98.0)

233

680

707

615

373

452

1902

552

Total
* 1-year

122

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Given the voluminous nature concerning the number of tables examining force use by
resistance type (Table 6-9 through 6-23), the next three tables offer a more consolidated look.
Table 6-24 combines Tables 6-9 through 6-13 in terms of the various types of force used in
relation to suspect encounters where the highest level of suspect resistance is failure to comply.
Tables 6-25 and 6-26 follow similar logic by combining physical defensive resistance (Tables 614 through 6-18) and physical aggressive resistance (Tables 6-19 through 6-23). Looking at
these three tables as a collective whole several themes emerge.
First, Columbus used the most soft hand tactics with all three types of resistance.
Second, Fort Wayne used the most hard hand tactics when dealing with all three types of
resistance. One explanation for this latter finding may be the fact that patrol officers in Fort
Wayne are not equipped with CEDs. Another explanation is that their policy may be considered
less restrictive than the six agencies that link resistance to force. However, Columbus also does
not link resistance to force and we do not find a similar result in that city. Third, Colorado
Springs CEDs most often when dealing with all three types of resistance. Perhaps the wheel
design helps explain such a finding, although from the text of the policy it seems rather apparent
that officers are discouraged from using a CED on a passive or verbally resistant suspect.
Fourth, Albuquerque was most reliant on chemical sprays when dealing with both
physically defensive and aggressive suspects. Fifth, Columbus officers were most likely to use
chemical sprays on failure to comply suspects, which would seem in accordance with their
policy (chemical sprays are placed low on the continuum policy). Sixth, Knoxville had the
highest baton usage on failure to comply suspects (9.4%), despite the fact that its use is restricted
within their policy.

123

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-24. Force Usage when Highest Level of Resistance was Failure to Comply
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Soft
Hands

25
(43.1)

270
(65.2)

5
(62.5)

57
(76.0)

222
(66.1)

24
(75.0)

880
(59.1)

34
(85.0)

Hard
Hands

11
(19.0)

29
(7.0)

0
(0.0)

17
(22.7)

13
(3.9)

2
(6.2)

24
(1.6)

1
(2.5)

Chemical
Spray

7
(12.1)

19
(4.6)

0
(0.0)

13
(17.3)

79
(23.5)

0
(0.0)

467
(31.4)

2
(5.0)

CED

26
(44.8)

80
(19.3)

3
(37.5)

0
(0.0)

23
(6.8)

7
(21.9)

110
(7.4)

5
(12.5)

Baton

0
(0.0)

1
(0.2)

0
(0.0)

1
(1.3)

0
(0.0)

3
(9.4)

2
(0.1)

0
(0.0)

Total

58

414

8

75

336

32

1489

40

* 1-year

124

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-25. Force Usage when Highest Level of Resistance was Physical Defensive
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Soft
Hands

199
(45.3)

607
(78.4)

466
(65.8)

1090
(85.2)

823
(73.7)

327
(82.0)

1848
(83.5)

621
(87.3)

Hard
Hands

97
(22.1)

101
(13.0)

104
(14.7)

405
(31.7)

70
(6.3)

61
(15.3)

77
(3.5)

106
(14.9)

Chemical
Spray

44
(10.0)

31
(4.0)

130
(18.4)

232
(18.1)

82
(7.3)

10
(2.5)

202
(9.1)

18
(2.5)

CED

220
(50.1)

172
(22.2)

132
(18.6)

5
(0.4)

267
(23.9)

69
(17.3)

217
(9.8)

75
(10.5)

Baton

1
(0.2)

1
(0.1)

10
(1.4)

26
(2.0)

2
(0.2)

6
(1.5)

8
(0.4)

3
(0.4)

Total

439

774

708

1279

1117

399

2213

711

* 1-year

125

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-26. Force Usage when Highest Level of Resistance was Physical Aggressive
Col. Sp.

Port.*

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Soft
Hands

93
(39.9)

439
(64.6)

424
(60.0)

500
(81.3)

292
(78.3)

395
(87.4)

1510
(79.4)

479
(86.8)

Hard
Hands

45
(19.3)

122
(17.9)

137
(19.4)

217
(35.3)

45
(12.1)

107
(23.7)

265
(13.9)

100
(18.1)

Chemical
Spray

40
(17.2)

90
(13.2)

227
(32.1)

190
(30.9)

40
(10.7)

32
(7.1)

337
(17.7)

35
(6.3)

CED

119
(51.1)

226
(33.2)

146
(20.7)

2
(0.3)

109
(29.2)

103
(22.8)

247
(13.0)

170
(30.8)

Baton

2
(0.9)

14
(2.1)

27
(3.8)

16
(2.6)

1
(0.3)

11
(2.4)

20
(1.1)

11
(2.0)

Total

23

680

707

615

373

452

1902

552

* 1-year

126

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
From this set of analyses it is difficult to state apparent policy effects. For instance,
Colorado Springs’ high frequency use of a CED, or Knoxville’s high frequency use of a baton,
on passive and verbally defensive suspects despite their policies advising against such usage; or
the extent to which Portland officers used hard hand tactics on defensively resistant suspects.
However, there are other findings that seem to be in line with an agency’s policy and the type of
force used in relation to the type of suspect resistance faced, as many of the force types used by
officers across all the cities fall in line or within the framework of what is permitted.

Multivariate Analyses
In addition to examining various forms of force in relation to suspect resistance in a
bivariate manner, we also wanted to look more globally at the severity of force officers use while
controlling for suspect resistance levels and other factors that may account for why officers use
force. As a result, we ran two multivariate models (ordered logit) with the highest level of force
as the dependent variable (0=soft hands, 1=hard hands and chemical spray, 2=CED and impact
weapons, 3=deadly force).43
Table 6-27 provides an overview of how the independent variables were defined, along
with hypothesized relationships to force. Table 6-28 provides descriptive statistics for the
43

Even with multivariate analyses, it is important to note the difficulty of isolating or pinpointing a so-called “policy
effect.” Relying on force reports for data analyses ultimately limits the amount of control. While the multivariate
approach allows us to control for numerous alternative factors that may cause officers to use certain types of force,
one cannot control all factors. For instance, we cannot control for multiple uses of the same resistance type within
the same encounter as noted in footnote 43. Nonetheless, our concerns are somewhat mitigated when we are
comparing cities. More specifically, one would have to assume that multiple instances of the same type of resistance
operates differently across cities for this to be a concern. Hence, if one city is found to be using significantly more
force, we can make a reasonable assumption it is a true effect rather than one attributed to multiple instances of the
same resistance. For the converse to be true, one would have to suspect that citizen resistance is somehow unique to
a given city.

127

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

dependent and each of the independent variables.44 The models include variables that have been
found to influence (or posited to influence) police use of force behavior in prior research
(Paoline and Terrill, 2004; Paoline and Terrill, 2007; Terrill and Mastrofski, 2002; Terrill,
Paoline, and Manning, 2003).45 For example, one would expect that officers may resort to force,
irrespective of explicit resistance, if a suspect has a weapon. Similarly, prior research has shown
suspect characteristics such as sex and race influence police behavior. Perhaps the key variables
in each of the models are the dummy variables reflecting the cities. By incorporating these
measures, we are most closely able to isolate policy effects across the cities.

44

Columbus did not capture suspect race as part of their force reporting system and is thus excluded from the
analyses.
45

We were somewhat limited with respect to the situational independent variables that could be used in the analyses
given that the official reports did not reliably capture numerous measures (e.g., suspect demeanor, suspect
wealth/class, or number of bystanders/officers present).

128

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-27. Description of Independent Variables
Variable

Hypothesized Effect

Definition

Suspect Resistance

+

4=deadly, 3=physical aggressive,
2=physical defensive, 1=failure to
comply, 0=none

Suspect Sex

+

1=male, 0=female

Suspect Race

+

1=non-white, 0=white

Suspect Age

+

Years

Suspect Drug

+

1=Suspect shows behavioral effects of
drug/alcohol, 0=all other

Suspect Mental

+

1=Suspect shows behavioral effects of
mental impairment, 0=all other

Suspect Weapon

+

1=Suspect has weapon, 0=all other

Colorado Springs

+/-

1=Colorado Springs, 0=all other

Charlotte-Mecklenburg

+/-

1=Charlotte-Mecklenburg, 0=all other

Portland

+/-

1=Portland, 0=all other

St. Petersburg

+/-

1=St. Petersburg, 0=all other

Knoxville

+/-

1=Knoxville, 0=all other

Albuquerque

+/-

1=Albuquerque, 0=all other

Fort Wayne

+/-

1=Fort Wayne, 0=all other

129

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-28. Descriptive Statistics for Multivariate Models
Variable

Range

Mean

Std. Dev.

Dependent
Highest level of force
Soft Hands
Hard Hands/Chemical Spray
CED/Impact Weapons
Deadly

0-3

.76

.833

Suspect Resistance

0-4

2.23

.739

Suspect Sex

0-1

.86

.345

Suspect Race

0-1

.61

.488

Suspect Age

8-84

30.42

11.150

Suspect Drug

0-1

.47

.499

Suspect Mental

0-1

.09

.292

Suspect Weapon

0-1

.10

.302

Colorado Springs

0-1

.07

.308

Charlotte-Mecklenburg

0-1

.13

.334

Portland

0-1

.20

.400

St. Petersburg

0-1

.18

.386

Knoxville

0-1

.09

.284

Albuquerque

0-1

.14

.347

Fort Wayne

0-1

.19

.394

Independent

130

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

We begin by examining the highest level of force, as illustrated in Table 6-29. In this
model, we use Albuquerque as the reference category since their policy may be considered less
restrictive than the other cities as outlined in Table 6-8 (e.g., with the exception of impact force,
all other forms of force are located at the bottom of the continuum); consequently we would
expect Albuquerque officers to use more force. Thus, we posit than officers in each of the cities
in the model will use less force than Albuquerque officers.
With the exception of Colorado Springs, the results show, as posited, that officers in each
of the cities are all less likely to use higher forms of force compared to Albuquerque officers. Of
these, only Portland fails to reach statistical significance. Conversely, Colorado Springs officers
were significantly more likely to use higher levels of force, compared to Albuquerque, most
likely the result of such high CED use as indicated in previous analyses. In general, such
findings lend some degree of support in terms of a policy effect (i.e., the policy translates to
behavioral differences).

131

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-29. Ordered Logit Estimates of Highest Force (N = 10,428)
Reference Category: Albuquerque
Variable

S.E.

Sig.

.557

.029

.000

Suspect Sex

1.105

.064

.000

Suspect Race

.047

.043

.279

Suspect Age

.006

.002

.000

Suspect Drug

- .151

.042

.000

Suspect Mental

.261

.069

.000

Suspect Weapon

.639

.065

.000

Colorado Springs

1.089

.091

.000

Charlotte-Mecklenburg

- .482

.073

.000

Portland

- .120

.069

.082

St. Petersburg

- .297

.071

.000

Knoxville

- .492

.084

.000

Fort Wayne

- .591

.069

.000

Suspect Resistance

Chi Square
- 2 Log Likelihood
Pseudo R square

B

1239.514
13160.279
.118

132

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-30. Ordered Logit Estimates of Highest Force (N = 10,428)
Reference Category: Charlotte-Mecklenburg
Variable

S.E.

Sig.

.557

.029

.000

Suspect Sex

1.105

.064

.000

Suspect Race

.047

.043

.279

Suspect Age

.006

.002

.000

Suspect Drug

- .151

.042

.000

Suspect Mental

.261

.069

.000

Suspect Weapon

.639

.065

.000

Colorado Springs

1.571

.094

.000

Albuquerque

.482

.073

.000

Portland

.362

.072

.000

St. Petersburg

.184

.074

.012

Knoxville

- .011

.086

.901

Fort Wayne

- .109

.072

.127

Suspect Resistance

Chi Square
- 2 Log Likelihood
Pseudo R square

B

1239.514
13160.279
.118

133

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-30 also examines the highest level of force. In this model we use CharlotteMecklenburg as the reference category since their policy may be considered more restrictive
than the other cities as outlined in Table 6-8 (e.g., CED and impact force are located at the top of
the continuum); consequently we would expect Charlotte-Mecklenburg officers to use less force.
Thus, we posit than officers in each of the cities in the model will use more force than CharlotteMecklenburg officers.
Here we find that officers in Colorado Springs, Albuquerque, Portland, and St.
Petersburg all use significantly more force than Charlotte-Mecklenburg officers. However, no
statistically significant differences were found between Knoxville, Fort Wayne, and CharlotteMecklenburg officers. As result, we once again find some partial support in terms of a policy
effect (i.e., the policy translates to behavioral differences).
Summary
Much like our summary assessment from the bivariate analyses, the multivariate analyses
indicate that it is difficult to pinpoint any sort of consistent policy effects. In some cases we find
officers operating along the lines that one might expect given the varying policy approaches,
while in other instances we do not. What appears most evident is that Colorado Springs officers
are using the highest levels of force net of policy and other controls (e.g., suspect resistance,
suspect having a weapon, suspect demographics).

134

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Suspect Injuries and Force
In this section we examine the extent to which officers reported suspects being injured.
In particular, we want to examine whether any differences in injuries may be related to the
varying types of force policies used by our eight agencies. We begin with Table 6-31. Figures
are computed with respect to the frequency of injury in relation to the total number of force
reports filed per city. As illustrated, Charlotte-Mecklenburg, by far, reported the highest number
of suspect injuries at 73.5 percent. This is followed by Knoxville (45.9%) and Colorado Springs
(37.1%). The next four cities (Fort Wayne, Portland, Albuquerque, and Columbus) all reported
in the 26 to 27 percent range. Finally, St. Petersburg reported the least amount of injuries at 15.9
percent.
Table 6-32 looks at the types of injuries suspects sustained. Note that no percentages are
offered for Colorado Springs nor Albuquerque since this information was not reliably captured
on their use of force report form. As shown, Fort Wayne had the most bruises (17.4%), Portland
the most abrasions (50.4%), Knoxville the most lacerations (49.2%), Fort Wayne the most
broken bones (2.2% , although we offer caution since the number of cases are so few) and Fort
Wayne the most “other” injuries.46 Interestingly, in three of the five categories Fort Wayne
ranked at the top. Further, the city with the most suspect injuries (Charlotte-Mecklenburg) did
not rank at the top in any individual injury type category.

46

Precisely how these categories were defined across all the cities is unknown. In some cities (e.g., Portland), these
five categories were pre-labeled on the force report form without definition and were simply circled by the officer.
In other cities (e.g., Columbus, Fort Wayne, St. Petersburg), an empty fill-in box was presented to the officer to
identify the injury. If the research team had to make a determination of injury type, we used the following
definitions: bruises (any mention of “bruise” by the officer or reporting of black and blue skin conditions), abrasions
(any mention of “abrasion” by the officer or reporting of scraps/surface to the skin marks), lacerations (any mention
of “laceration” by the officer or reporting of cuts to the skin), broken bones (any mention of “broken bone” by the
officer or reporting of a fracture).

135

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-31. Was Suspect Injured?
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

274
(37.1%)

1206
(27.6%)

392
(26.9%)

542
(27.1%)

302
(15.9%)

425
(45.9%)

1531
(26.4%)

979
(73.5%)

No

464
(62.9%)

3168
(72.4%)

1066
(73.1%)

1455
(72.9%)

1595
(84.1%)

501
(54.1%)

4256
(73.6%)

353
(26.5%)

738

4374

1458

1997

1897

926

5787

1332

Total

136

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-32. Suspect Injury Types
Col. Sp.

Portl.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Bruises

-

158
(13.1%)

-

93
(17.4%)

15
(5.0%)

18
(4.3%)

64
(4.4%)

18
(1.8%)

Abrasions

-

608
(50.4%)

-

139
(26.0%)

140
(47.0%)

150
(35.5%)

467
(32.1%)

353
(36.1%)

Lacerations

-

195
(16.2%)

-

126
(23.6%)

93
(31.2%)

208
(49.2%)

674
(46.3%)

404
(41.3%)

Broken Bones

-

6
(0.5%)

-

12
(2.2%)

5
(1.7%)

8
(1.9%)

18
(1.2%)

10
(1.0%)

Other

-

239
(19.8%)

-

165
(30.8%)

45
(15.1%)

39
(9.2%)

231
(16.0%)

194
(19.8%)

Total

-

1206

-

535

298

423

1454

979

137

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Several interesting points of reference emerge. First, Charlotte-Mecklenburg and St.
Petersburg both restrict CED use to aggressively resistant suspects (high on the continuum), yet
they fall on opposite ends in terms of suspect injuries (Charlotte-Mecklenburg reported the most
injuries while St. Petersburg the least). Moreover, Colorado Springs officers used a CED far
more often than officers in all the other cities, yet they have the third highest percentage of
injuries. Second, arguably the three cities with the least restrictive policies (Fort Wayne and
Columbus because they do not link resistance to force, and Albuquerque because they are
permitted to use any form of force on any form of resistance with the exception of impact force
on failure to comply suspects) all fall in the middle to low range on suspect injuries. One may
posit that less policy restriction may lead to more injuries (i.e., because officers will have more
freedom), but in fact we find that such policies were related with less injuries comparatively.
Third, Knoxville and Colorado Springs resulted in the second and third highest
percentage of injuries. However, their policies are substantially different. Colorado Springs
employs a wheel design and reserves hard hand tactics and impact force to aggressive resistant
suspects, while Knoxville allows officers to use not only hard hands tactics and impact methods
on defensively resistant suspects, but also chemical spray and a CED. Fourth, if suspect injuries
are ranked in terms of most injuries to least injuries and then compared to suspect resistance
levels (see Table 6-7b) there is a fair amount of similarity found. For example, St. Petersburg
reports the lowest mean level of resistance and the lowest percentage of suspects being injured.
The biggest difference when making this type of comparison is found in Albuquerque, which
ranks highest in terms of resistance levels, but just sixth in injuries.

138

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Fifth, if suspect injuries are ranked in terms of most injuries to least and then compared
to officer force levels (see Table 6-7a), there is less similarity found than when compared to
resistance levels. For instance, Charlotte-Mecklenburg had the most injuries but only the fourth
highest level of force, while Albuquerque had just the sixth most injuries but the second highest
level of force.

Officer Injuries and Force
In this section we examine the extent to which officers reported receiving injuries during
their force encounters with suspects. According to the 2007 Federal Bureau of Investigation’s
publication of Law Enforcement Officers Killed and Assaulted, 57 officers across the United
States were killed by citizens, and 59,201 were assaulted (rate of 11.4 per 100 officers), of which
5,479 resulted in a documental injury (rate of 3.0 per 100 officers). Our figures are computed
with respect to the frequency of injury in relation to the total number of force reports filed per
city. Note that no percentages are offered for Albuquerque since this information was not
reliably captured on their use of force report form. As shown in Table 6-33, Knoxville reported
the highest number of officer injuries (14.8%), followed by Charlotte-Mecklenburg (13.4%),
Colorado Springs (12.7%), Fort Wayne (12.2%), St. Petersburg (9.3%), Portland (9.0%), and
Columbus (8.1%).
Table 6-34 looks at the types of injuries officers sustained. No percentages are offered
for Colorado Springs nor Albuquerque as this information was not reliably captured on their
force report form. As shown, Portland had the most bruises (16.5%), Charlotte-Mecklenburg the
most abrasions (43.8%), Knoxville the most lacerations (33.6%) and broken bones (2.9% ,

139

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

although we offer caution since the number of cases are so few), and Columbus the most “other”
injuries.47 Interestingly, in two of the five categories Knoxville ranked at the top.

47

Similar to suspect injuries, precisely how these categories were defined across all the cities is unknown. In some
cities these categories were pre-labeled on the force report. If the research team had to make a determination,we
used the following: bruises (any mention of “bruise” or reporting of black and blue skin conditions), abrasions (any
mention of “abrasion” or reporting of scraps/surface to the skin marks), lacerations (any mention of “laceration” or
reporting of cuts to the skin), broken bones (any mention of “broken bone” or reporting of a fracture).

140

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-33. Was Officer Injured?
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Yes

94
(12.7%)

393
(9.0%)

-

243
(12.2%)

177
(9.3%)

137
(14.8%)

467
(8.1%)

178
(13.4%)

No

644
(87.3%)

3981
(91.0%)

-

1754
(87.8%)

1720
(90.7%)

789
(85.2%)

5320
(91.9%)

1154
(86.6%)

738

4374

-

1997

1897

926

5787

1332

Total

141

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 6-34. Officer Injury Types

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Bruises

-

65
(16.5%)

-

38
(15.8%)

24
(13.8%)

16
(11.7%)

30
(6.8%)

6
(3.4%)

Abrasions

-

170
(43.3%)

-

82
(34.0%)

64
(36.8%)

35
(25.5%)

110
(24.9%)

78
(43.8%)

Lacerations

-

55
(14.0%)

-

53
(22.0%)

37
(21.3%)

46
(33.6%)

126
(28.5%)

42
(23.6%)

Broken Bones

-

6
(1.5%)

-

3
(1.2%)

2
(1.1%)

4
(2.9%)

5
(1.1%)

3
(1.7%)

Other

-

97
(24.7%)

-

65
(27.0%)

47
(27.0%)

36
(26.3%)

171
(38.7%)

49
(27.5%)

Total

-

393

-

241

174

137

442

178

142

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Several interesting points of reference emerge. First, the percentages of officer injuries
are much lower on average than that of suspect injuries. In fact, the city with the highest
percentage of officer injuries (Knoxville, 14.8%), was still lower than the city with the lowest
percentage of suspect injuries (St. Petersburg, 15.9%). Thus, the extent to which officers are
injured, across all cities, is substantially lower than the extent to which suspected are injured.
Second, there is not a great deal of variation found across officer injuries when compared to
suspect injuries. The small range of injuries goes from 8.1 percent in Columbus up to just 14.8
percent in Knoxville. Recall that the range for suspect injuries was 15.9 percent in St.
Petersburg up to 73.5 percent in Charlotte-Mecklenburg. Third, Charlotte-Mecklenburg and St.
Petersburg both restrict CED use to aggressively resistant suspects (high on the force
continuum), yet Charlotte-Mecklenburg reported the second highest percentage of officer
injuries, while St. Petersburg reported just the fifth highest percentage out of seven cities.
Moreover, Colorado Springs officers used a CED far more frequently than officers in all the
other cities, yet this city still had the third highest percentage of officer injuries.
Fourth, two of the cities with arguably the least restrictive policies (Fort Wayne and
Columbus because they do not link resistance to force) fall in the middle to low range on officer
injuries. One may posit that less restrictive policies may lead to more injuries (i.e., because
officers will have more freedom), but in fact we find that such policies were related to less
injuries comparatively. Fifth, Knoxville and Charlotte-Mecklenburg resulted in the first and
second highest percentage of injuries. However, their policies are substantially different. In
particular, officers in Charlotte-Mecklenburg are restricted from using a CED or impact force on
suspects unless they are aggressively resistance. Knoxville’s policy is less restrictive with the

143

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

use of a CED and impact weapons, permitting such force to be used on suspects who display just
defensive resistance. Sixth, if officer injuries are ranked in terms of most injuries to least
injuries and then compared to suspect resistance levels (see Table 6-7b) there is a fair amount of
similarity found. For example, Knoxville reports the second highest mean level of resistance and
the highest percentage of officers being injured. Other cities show similarities in terms of degree
(i.e., the more resistance posed by suspects the more officers reported being injured).

144

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 7
Citizen Complaint Data
As with force reports, a key part of the second phase of the project involved the
collection of complaint data. In this chapter we examine complaints to investigate the
relationship between policy types and the number, types, and outcomes of citizen complaints.

Methodology
The data collection process for complaints mirrored that of the force data and consisted
of the same 24 month periods for each city. These time frames ensured that there was a
consistent two-year period with no changes in substance or reporting with respect to each site’s
use of force policies. Much like the force report data, some of the research sites coded all
complaint data electronically (i.e., Columbus, Charlotte-Mecklenburg, Colorado Springs, and
Portland), others only maintained paper files (i.e., Fort Wayne, Knoxville, and St. Petersburg),
and the others had a combination of electronic and paper data (i.e., Albuquerque). Sites that only
kept paper files of complaint cases required coding into electronic format. Even in sites that kept
data electronically, the research team had to refer to hard copies of the complaint files in order to
obtain the necessary information (e.g., dispositions and discipline). In some instances, this
meant having to gather the data from multiple, external sources. For example, in Albuquerque,
much of the complaint information was recorded and stored electronically by Internal Affairs.
However, certain key variables, such as the demographic information of complainants, were
better collected by the city’s Independent Review Office, the office that oversaw complaint
investigations. Thus, we had to collect complaint data from each source in order to adequately
capture all of the variables necessary for comparing complaints across each of the eight sites.
145

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

We wanted to ensure that the complaint data collected were comparable across sites.
This proved to be challenging because the complaint processes for each site varied. For
example, complaints were investigated solely by Internal Affairs in some sites (i.e., Fort Wayne),
by a combination of Internal Affairs and officers’ chain of command in other sites (i.e.,
Columbus, Charlotte-Mecklenburg, Colorado Springs, St. Petersburg, and Knoxville), or by a
combination of departmental entities and external oversight agencies in the remaining sites (i.e.,
Portland and Albuquerque). Furthermore, the process by which complaints were evaluated for
investigative merit also varied. In order to ensure that we were including similar types of
complaints across sites, only complaints that were investigated by some type of department or
external entity were included for analysis. As such, complaints that were rejected, declined,
inactivated, not deemed to be a violation of policy, mediated, withdrawn, resolved at intake, or
cancelled for cause were not included in the study.
We focused our attention on citizen complaints against officers. Our aim was to gain a
better understanding of the types, numbers, and outcomes of complaints filed in each
department. In all, we were able to collect all investigated citizen complaints for our eight sites.
Finally, it should be noted that, for the following analyses, only external complaints lodged
against sworn officers were included. As such, complaints against non-sworn police personnel
were not included in the study.
Across each of our research sites, a number of characteristics related to the complaint
incident, the citizens who filed the complaint, and the officers receiving the complaint were
collected. For the current purpose, however, the key variable that allows for comparisons to be
made across sites, and allows for the findings to be related back to each site’s force policy, was

146

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the type of alleged misconduct lodged against officers. Specifically, our focus centered on two
types of allegations that prior research has identified as important indicators when examining
complaints in relation to use of force: force and discourtesy allegations (Terrill and McCluskey,
2002). Across each site, each allegation of force and each allegation of discourtesy was coded
from the complaint data.
In addition to coding force and discourtesy complaints, common measures of the
investigation findings were coded as well. Four primary findings were coded from the data and
classified as sustained, exonerated, not sustained, or unfounded. Finally, common measures of
disciplinary actions were also created. Disciplinary actions were coded as none, supervisory
actions, command counseling, written reprimand, suspension, termination, or “other.” The
“other” category reflected actions taken infrequently, such as demotions or actions that were
pending at the time the data was collected.

Analyses and Findings
Frequency of Complaints
Our analysis begins by examining the frequency of citizen complaints across the eight
departments. Table 7-1 presents the frequencies and percentages of all citizen complaints across
each site. It should be noted that the analyses presented in this chapter focus on the number of
complaint allegations made against officers in the eight departments. In some cases, multiple
complaint allegations may have been made against officers in a single incident, while in other
instances multiple complaint allegations may have been made against multiple officers in an

147

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

incident. We treat each individual allegation of misconduct as an outcome of interest. In other
words, complaints refer to each separate allegation of misconduct filed against officers.
Based on Table 7-1, the total number of citizen complaints ranged from 94 in Fort Wayne
to 2,542 in Columbus for the two year period. Based on the total number of citizen complaints
filed, Fort Wayne, Charlotte-Mecklenburg, and Knoxville had the highest percentages of use of
force allegations. For example, of the 94 citizen complaints in Fort Wayne, 43 percent were for
use of force. Furthermore, just over a third of all external complaints were for use of force in
Charlotte-Mecklenburg, while almost one-quarter of all external complaints were for force in
Knoxville. St. Petersburg and Albuquerque were the two sites with the lowest number of
external force complaints with only 8.2 and 14.5 percent, respectively. Similar results were
found for external discourtesy complaints, Fort Wayne and Knoxville had the highest
percentages, while Albuquerque and St. Petersburg had the lowest percentages.
An interesting finding emerges when combining the total number of external force and
discourtesy complaints and examining them in relation to the total number of all external
complaints for each site. In Fort Wayne, 88 of the 94 (93.6%) external complaints were either
for alleged force or discourtesy violations. This is substantially higher than the next highest
departments, Knoxville (55%) and Charlotte-Mecklenburg (53%). Although Fort Wayne had the
lowest number of citizen complaints, the vast majority of these were for force or discourtesy
violations.

148

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The aim of this introductory section was to report the total number of citizen complaints
(i.e., allegations), as well as the number (and percentage of overall total) of use of force and
discourtesy allegations. While we find substantial variation (especially in totals) across our
cities, a couple of clear patterns emerge. First, in relation to the total citizen complaints levied
across our agencies, the percentage of use of force and discourtesy complaints was, by far, the
highest in Fort Wayne, and to a lesser extent in Knoxville and Charlotte-Mecklenburg (see Table
7-1). Second, on the other end of the spectrum, we found both St. Petersburg and Albuquerque which reported the lowest percentage of force and discourtesy complaints, when compared to all
citizen complaints.

149

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-1. Citizen Complaint Allegations: Overall, Use of Force, and Discourtesy

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

848

1350

962

94

245

158

390

2542

Total Citizen Use of
Force Complaints

192
(22.6%)

274
(20.3%)

139
(14.5%)

40
(42.5%)

20
(8.2%)

37
(23.4%)

132
(33.8%)

505
(19.9%)

Total Citizen
Discourtesy
Complaints

197
(23.2%)

244
(18.1%)

77
(8.0%)

48
(51.1%)

39
(15.9%)

50
(31.6%)

76
(19.5%)

600
(23.6%)

Total Citizen
Complaints

150

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Dispositions of Complaints
In addition to the types and frequencies of complaints levied against the eight
departments, we also collected data on complaint findings. Although allegations of misconduct
themselves, especially excessive/improper force, present serious problems for police
departments, accusations do not necessarily mean that the misconduct occurred. That is, even
though all complaints were deemed to have investigative merit by departments, complaints are
often difficult to substantiate (Lersch and Mieczkowski, 2000). Therefore, it is also important to
examine the dispositions of complaints to determine how often complaints, particularly those for
excessive/improper force and discourtesy, were substantiated across each department over the
two year period.
Our analysis of complaint dispositions starts with the total number of dispositions for all
types of citizen complaints investigated by departments. Table 7-2 presents the frequencies and
percentages for the four common measures of dispositions across departments: sustained (i.e.,
the misconduct occurred), not sustained (i.e., the misconduct could not be proven or disproven),
exonerated (i.e., the conduct in question was proper), or unfounded (the allegation of misconduct
was found to be false).48 These numbers provide a baseline for the results of complaint
investigations across departments.

48

Column totals in Table 7-2 may not add to 100 percent. This is due to the fact that there may have been findings
other than the four presented across sites. These were findings that were unique and often used infrequently by
departments or findings that were still pending at the time of the study. Also, findings in Portland are unique which
makes it difficult to make comparisons to the other seven sites. For example, almost 25 percent of complaint
dispositions in Portland were still pending at the time of the study. In addition, almost 45 percent of findings were
disposed of as service complaints, or “minor rule violations resolved by a supervisor.” This was unique to Portland
in the sense that service complaints were treated as dispositions, while in other sites these types of complaints would
have been addressed by the type of actions taken as a result of the investigation (e.g., disciplinary actions). These
aspects should be kept in mind when interpreting dispositions for Portland.

151

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-2. Total Investigated Citizen Complaint Dispositions

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

848

1350

962

94

245

158

390

2542

89 (10.5%)

29
(2.1%)

170
(17.7%)

9
(9.6%)

72
(29.4%)

22
(13.9%)

55
(14.1%)

226
(8.9%)

Not Sustained

108
(12.7%)

55
(4.1%)

217
(22.6%)

61
(64.9%)

106
(43.3%)

21
(13.3%)

281
(72.1%)

568
(22.3%)

Exonerated

273
(32.2%)

146
(10.8%)

431
(44.8%)

1
(1.1%)

15
(6.1%)

51
(32.3%)

28
(7.2%)

589
(23.2%)

Unfounded

355
(41.9%)

91
(6.7%)

144
(15.0%)

10
(10.6%)

35
(14.3%)

64
(40.5%)

26
(6.7%)

1147
(45.1%)

Total Citizen
Complaints
Sustained

152

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-3. Use of Force and Discourtesy Investigated Citizen Complaint Dispositions

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

192

274

139

40

20

37

132

505

Sustained

1
(0.5%)

4
(1.5%)

7
(5.0%)

0
(0.0%)

0
(0.0%)

1
(2.7%)

5
(3.8%)

5
(1.0%)

Not Sustained

22
(11.5%)

12
(4.4%)

25
(18.0%)

29
(72.5%)

6
(30.0%)

6
(16.2%)

114
(86.4%)

74
(14.7%)

Exonerated

88
(45.8%)

72
(26.3%)

96
(69.1%)

1
(2.5%)

4
(20.0%)

23
(62.2%)

7
(5.3%)

168
(33.3%)

Unfounded

76
(39.6%)

21
(7.7%)

11
(7.9%)

3
(7.5%)

4
(20.0%)

7
(8.9%)

6
(4.5%)

258
(51.1%)

197

244

77

48

39

50

76

600

Sustained

16
(8.1%)

5
(2.0%)

11
(14.3%)

6
(12.5%)

7
(17.9%)

14
(28.0%)

15
(19.7%)

40
(6.7%)

Not Sustained

39
(19.8%)

12
(4.9%)

36
(46.8%)

29
(60.4%)

28
(71.8%)

8
(16.0%)

54
(71.1%)

242
(40.3%)

Exonerated

48
(24.4%)

8
(3.3%)

13
(16.9%)

0
(0.0%)

0
(0.0%)

11
(22.0%)

5
(6.6%)

37
(6.2%)

Unfounded

92
(46.7%)

11
(4.5%)

17
(22.1%)

7
(14.6%)

4
(10.3%)

17
(34.0%)

2
(2.6%)

281
(46.8%)

Use of Force
Complaints

Discourtesy
Complaints

153

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Although the results in Table 7-2 display frequencies and percentages for each finding category,
the current discussion focuses on the number of sustained citizen complaints across each site.
Based on the total number of citizen complaints, the number of sustained dispositions ranged
from approximately 10 percent, of the total number of complaints, in Fort Wayne to almost 30
percent in St. Petersburg.
Table 7-3 presents dispositions for citizen complaints for the use of force and
discourtesy. Recall from Table 7-1 that the frequency of force complaints roughly ranged
between 8 and 43 percent of the total number of citizen complaints lodged across departments.
Based upon the results in Table 7-3, relatively few of these force complaints are sustained
against officers from any of the eight sites. In fact, the percentage of external force complaints
sustained ranged from zero (in Fort Wayne and St. Petersburg) to only 5 percent (in
Albuquerque).
Table 7-3 also displays dispositions for discourtesy complaints. When compared to use
of force dispositions, sustained dispositions for discourtesy were found to be much more
frequent. Recall from Table 7-1 that roughly 8 (Albuquerque) to 51 (Fort Wayne) percent of
departments’ complaints were for discourtesy. The total number of sustained discourtesy
complaints ranged from approximately 8 (Colorado Springs) to 28 (Knoxville) percent.49
Summary
The focus of this section was to assess complaint dispositions (i.e., sustained versus all
others) across our eight cities. As we examine overall, use of force, and discourtesy allegations,
a few interesting findings emerge. First, with respect to all citizen complaints combined, St.
Petersburg (29.4) and Albuquerque (17.7) had the highest percentages of sustained complaints,
49

Again, this discussion excludes Portland due to the nature of their findings.

154

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

while Fort Wayne (9.6) and Columbus (8.9) had the lowest percentage. Recall that there were
difficulties in comparing Portland’s percentages to the rest of the cities.
In examining use of force and discourtesy allegations, it is clear that a much smaller
percentage of force complaints are sustained. In illustrating this point, we found that the highest
percentage of total sustained use of force complaints were found in Albuquerque (5.0%) and
Charlotte-Mecklenburg (3.8%). So, at the high end, we find that only one out of every twenty
use of force complaints is sustained. On the other hand, a higher percentage of sustained
complaints were found for discourtesy allegations. Knoxville was found to have the highest
percentage of sustained allegations, while Columbus had the fewest.

Disciplinary Actions
We also examined, across the eight cities, the types of actions taken against officers
accused of misconduct. These results are presented in Tables 7-4 through 7-6. We begin by
examining the various actions taken across all types of citizen complaints for each city, which is
reported in Table 7-4. The first row indicates the total number of actions taken for all complaint
types. Two aspects should be noted about the coding of disciplinary actions. First, the total
number of actions could exceed the total number of complaints. This is due to the fact that
multiple actions could be taken for a single complaint. For example, an officer could receive a
written reprimand and a suspension for a single complaint. Furthermore, actions may have been
taken even if complaints were not found to be substantiated. Second, disciplinary actions were
usually administered based on the entire complaint investigation, meaning that in cases where
multiple allegations were made against officers, discipline was based upon the totality of the

155

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

circumstances of the complaint case. Since our analyses focus on each separate complaint
allegation, the same disciplinary action was coded for each separate complaint allegation. Thus,
some of the frequencies presented may initially appear rather high. For example, there were 20
suspensions coded in Albuquerque over the two year period. This was due to the fact that the
same disciplinary action would have been coded for each allegation associated with the
complaint incident. These two issues should be kept in mind when assessing discipline across
each city.
Most citizen complaints resulted in no disciplinary action being taken. Across cities,
roughly 71 (in Colorado Springs and St. Petersburg) to 92 (in Columbus) percent of complaints
resulted in no action. This is not surprising given that the majority of complaints against officers
were not substantiated. Across cities, there were five primary types of actions taken.
Supervisory actions consisted of counseling and training at the officers’ direct supervisory level.
Supervisory actions were considered to be informal and most cities did not consider these to be
disciplinary in nature. The remaining actions were formal actions that consisted of command
counseling, written reprimands, suspensions, and terminations. One way to make comparisons is
to examine the nature of discipline for complaints where at least one type of action was taken.
Although we present all types of actions, discussion is focused on comparisons across cities for
the formal disciplinary actions of written reprimands, suspensions, and terminations.50
For all complaint types, the results in Table 7-4 indicated that when at least some type of
action was taken, more formal forms of discipline were taken in St. Petersburg and Charlotte50

We restrict discussion to these three categories for two primary reasons. First, supervisory actions were not
considered to be formal discipline. Second, it was difficult to identify the nature of command counseling for some
cities. For example, in Colorado Springs, counseling was considered to be an informal supervisory action. The
nature of counseling in Charlotte-Mecklenburg, on the other hand (i.e., formal or informal discipline), was not as
discernable. To ensure we are making consistent comparisons for similar types of disciplinary actions, we focus on
written reprimands, suspensions, and terminations, all of which were treated the same across sites.

156

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Mecklenburg. For example, in St. Petersburg some type of disciplinary action was taken in 72
out of the 245 total actions.51 Approximately 53 percent of these 72 actions were formal
disciplinary measures consisting of written reprimands, suspensions, or terminations. For
Charlotte-Mecklenburg 51 percent of their actions were formal in nature. On the other hand, the
results in Table 7-4 indicate that Colorado Springs and Portland relied on more informal actions
when examining all types of citizen complaints.
With respect to disciplinary actions for use of force citizen complaints, a few patterns
emerge. As Table 7-5 illustrates, once again, formal actions were extremely infrequent and this
is due to the fact that few use of force citizen complaints were substantiated across sites. Once
again, Portland and Colorado Springs relied on informal actions more often than the other cities.
In terms of formal disciplinary actions, Charlotte-Mecklenburg (3.1) and Knoxville (2.7)
recorded the highest percentages.
Turning next to disciplinary actions for discourtesy complaints (see Table 7-6), formal
actions were taken infrequently, but at consistently higher amounts compared to the use of force
complaints. Portland, Colorado Springs, and Knoxville relied on informal actions to address
discourtesy complaints more than the other five cities. On the other hand, CharlotteMecklenburg had the highest percentage of formal actions and suspensions for discourtesy,
while Columbus did not take a single formal action for any discourtesy citizen complaint.

51

Calculated by subtracting the no action taken row from the total number of disciplinary action row.

157

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-4. Investigated Citizen Complaint Disciplinary Actions: All Complaints

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Disciplinary Actions for
All Citizen Complaints

919

1350

996

94

245

159

400

2543

No Action
Taken

649
(70.6%)

236
(17.5%)

790
(79.3%)

82
(87.2%)

173
(70.6%)

133
(83.6%)

324
(81.0%)

2333
(91.8%)

Supervisory
Action

204
(22.2%)

711
(52.7%)

111
(11.1%)

1
(1.1%)

0
(0.0%)

17
(10.7%)

1
(0.2%)

4
(0.2%)

Command
Counseling

N/A

2
(0.1%)

N/A

N/A

32
(13.1%)

N/A

34
(8.5%)

160
(6.3%)

Written
Reprimand

19
(2.1%)

8
(0.6%)

36
(3.6%)

4
(4.3%)

22
(9.0%)

1
(0.6%)

16
(4.0%)

8
(0.3%)

Suspension

3
(0.3%)

5
(0.4%)

20
(2.0%)

1
(1.1%)

12
(4.9%)

7
(4.4%)

22
(5.5%)

16
(0.6%)

Termination

3
(0.3%)

0
(0.0%)

3
(0.3%)

0
(0.0%)

4
(1.6%)

0
(0.0%)

1
(0.2%)

22
(0.9%)

Other

41
(4.5%)

388
(28.7%)

36
(3.6%)

6
(6.4%)

2
(0.8%)

1
(0.6%)

2
(0.5%)

0
(0.0%)

158

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-5. Investigated Citizen Complaint Disciplinary Actions: Use of Force Complaints

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Disciplinary Actions for
Use of Force
Complaints

199

274

139

40

20

37

132

505

No Action
Taken

159
(79.9%)

84
(30.7%)

130
(93.5%)

39
(97.5%)

20
(100.0%)

36
(97.3%)

126
(95.5%)

500
(99.0%)

Supervisory
Action

35
(17.6%)

68
(24.8%)

4
(2.9%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

2
(0.4%)

Command
Counseling

N/A

2
(0.7%)

N/A

N/A

0
(0.0%)

N/A

1
(0.8.%)

N/A

Written
Reprimand

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

3
(2.3%)

2
(0.4%)

Suspension

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

1
(2.7%)

1
(0.8.%)

1
(0.2%)

Termination

0
(0.0%)

0
(0.0%)

1
(0.7%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

Other

5
(2.5%)

120
(43.8%)

4
(2.9%)

1
(2.5%)

0
(0.0%)

0
(0.0%)

1
(0.8%)

0
(0.0%)

159

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-6. Investigated Citizen Complaint Disciplinary Actions: Discourtesy Complaints
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Disciplinary Actions for
Discourtesy Complaints

218

244

77

48

39

51

81

600

No Action
Taken

157
(72.0%)

24
(9.8%)

65
(84.4%)

40
(83.3%)

32
(82.1%)

35
(68.6%)

59
(72.8%)

561
(93.5%)

Supervisory
Action

54
(24.8%)

155
(63.5%)

4
(5.2%)

1
(2.1%)

0
(0.0%)

12
(23.5%)

0
(0.0%)

0
(0.0%)

Command
Counseling

N/A

0
(0.0%)

N/A

N/A

3
(7.7%)

N/A

12
(14.8%)

39
(6.5%)

Written
Reprimand

3
(1.4%)

3
(1.2%)

6
(7.8%)

3
(6.2%)

2
(5.1%)

1
(2.0%)

4
(4.9%)

0
(0.0%)

Suspension

0
(0.0%)

0
(0.0%)

1
(1.3%)

1
(2.1%)

1
(2.6%)

2
(3.9%)

6
(7.4%)

0
(0.0%)

Termination

0
(0.0%)

0
(0.0%)

1
(1.3%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

Other

4
(1.8%)

62
(25.4%)

0
(0.0%)

3
(6.2%)

1
(2.6%)

1
(2.0%)

0
(0.0%)

0
(0.0%)

160

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The goal of this section was to shed light on the disciplinary actions across total
citizen complaints, as well as those for use of force and discourtesy. One thing that is
abundantly clear is that formal disciplinary action (i.e., written reprimands, suspensions, and
terminations), across all types of complaints, is the exception rather than the norm. When we
compare across the cities, we find a few common tendencies. More specifically, across all
allegations, and those for force and discourtesy, Colorado Springs and Portland (and Knoxville
for discourtesy complaints only) consistently relied on formal disciplinary actions less frequently
than all other cities. The cities that did lead the way, in terms of formal disciplinary actions
across complaint types, were St. Petersburg (for all citizen complaints combined), Albuquerque
(for discourtesy allegations) and Charlotte-Mecklenburg (for all complaints combined as well as
use of force and discourtesy allegations).

Frequency of Complaints by Workload Measures
While the preceding sections were concerned with describing the frequency of force and
discourtesy complaints (and subsequent outcomes) across cities, in making comparisons to
overall complaint totals, the remaining sections focus on more in-depth analyses of our primary
areas of concern - use of force and discourtesy allegations. We begin this discussion by
examining how often external use of force and discourtesy complaints were made against the
eight cities in relation to four department workload measures: calls for service, reported crimes,
arrests, and use of force reports. Table 7-7 reports these results for use of force complaints,
while Table 7-8 presents the results for discourtesy complaints. Comparing use of force

161

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

complaints by calls for service, we find substantial variation across the eight cities. For example,
in Portland, citizens filed one force complaint for every 1,631 calls for service. In contrast, use
of force complaints in St. Petersburg were filed by citizens once for every 17,885 calls for
service. Thus, when using calls for service as a baseline, Portland received roughly ten times
more force complaints than officers in St. Petersburg. With respect to discourtesy complaints,
similar patterns were found for Portland and Charlotte-Mecklenburg. Portland had a
substantially higher discourtesy complaint rate per 1,000 calls for service then the other seven
cities. Charlotte-Mecklenburg, on the other hand, had the lowest rate: one discourtesy complaint
for every 10,663 calls for service.
Variation was also present when comparing each city’s complaint rates by Part I reported
crimes. Colorado Springs had the highest rates of force and discourtesy complaints per 1,000
Part I crimes across the eight cities. Citizens filed use of force complaints for every 202 Part I
crimes and filed discourtesy complaints for every 197 Part I crimes. On the other hand, St.
Petersburg had use of force and discourtesy complaints filed once for every 1,959 and 1,004 Part
I crimes, respectively. In the context of the amount of crime, officers in Colorado Springs
received use of force complaints approximately ten times more frequently, and discourtesy
complaints roughly five times more frequently, than officers in St. Petersburg.
When looking at the frequency of use of force and discourtesy complaints in relation to
the number of arrests for Part I crimes, officers in Columbus had the highest rates. Columbus
had a use of force complaint filed for every 27 arrests and a discourtesy complaint filed for every
23 arrests. Once again, St. Petersburg had the lowest complaint rates in relation to arrests. St.
Petersburg had a use of force complaint filed for every 314 arrests and a discourtesy complaint

162

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

filed for every 161 arrests. In the context of the number of arrests, officers in Columbus received
external complaints approximately twelve (use of force) and seven (discourtesy) times more
frequently than officers in St. Petersburg.
Finally, we examined the frequency of use of force and discourtesy complaints in relation
to the number of use of force reports filed in each city over the two year period. When
examining the results in Tables 7-7 and 7-8, several interesting patterns were found. For five of
the eight cities, the number of force and discourtesy complaints per force report was calculated
to be one complaint for approximately every 10 to 25 force reports filed. Two cities, however,
had much lower complaint rates. First, Fort Wayne had one force complaint for every 50 force
reports filed, and one discourtesy complaint for every 42 force complaints filed. Second, St.
Petersburg had one force complaint for every 95 force reports filed, and one discourtesy
complaint for every 49 force reports filed. On the other hand, Colorado Springs had higher
complaint rates than the other cities. In Colorado Springs, citizens filed one use of force
complaint for every four force reports. This was also the same for discourtesy complaints, one
complaint for every four force reports filed. Thus, Colorado Springs had approximately use of
force complaints filed almost 24 times more frequently and discourtesy complaints filed almost
12 times more frequently than St. Petersburg (the city with the lowest rate).

163

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-7. External Use of Force Complaint Comparison by
Calls for Service, Reported Crimes, Arrests, and Force Reports

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Complaint/Service
Calls (1,000)

0.35

0.61

0.21

0.11

0.06

0.08

0.16

0.33

Complaint per call

1 of 2,803

1 of 1,631

1 of 4,693

1 of 8,854

1 of 17,885

1 of 12,058

1 of 6,139

1 of 3,026

4.94

3.79

1.97

1.72

0.51

1.39

1.14

4.28

1 of 202

1 of 264

1 of 508

1 of 580

1 of 1,959

1 of 720

1 of 876

1 of 233

Complaint/Part I
Arrests (1,000)

21.70

19.10

22.00

8.08

3.18

6.18

10.94

37.30

Complaint per arrest

1 of 46

1 of 52

1 of 45

1 of 124

1 of 314

1 of 162

1 of 91

1 of 27

Complaint/Force
Report (100)

26.01

6.26

9.50

2.00

1.05

3.99

9.91

8.70

Complaint per force
report

1 of 4

1 of 16

1 of 10

1 of 50

1 of 95

1 of 25

1 of 10

1 of 11

192

274

139

40

20

37

132

505

Complaint/Part I
Crimes (1,000)
Complaint per crime

Total Number
External Force
Complaints

164

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-8. External Discourtesy Allegation Comparison by
Calls for Service, Reported Crimes, Arrests, and Force Reports

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Complaint/Service
Calls (1,000)

0.36

0.55

0.12

0.14

0.11

0.11

0.09

0.39

Complaint per call

1 of 2,732

1 of 1,831

1 of 8,472

1 of 7,378

1 of 9,172

1 of 8,923

1 of 10,663

1 of 2,547

5.07

3.37

1.09

2.07

0.99

1.88

0.66

5.08

1 of 197

1 of 296

1 of 918

1 of 483

1 of 1,004

1 of 533

1 of 1,522

1 of 197

Complaint/Part I
Arrests (1,000)

22.24

17.05

12.21

9.70

6.21

8.35

6.30

44.35

Complaint per arrest

1 of 45

1 of 59

1 of 82

1 of 103

1 of 161

1 of 120

1 of 159

1 of 23

Complaint/Force
Report (100)

26.69

5.58

5.28

2.40

2.06

5.40

5.70

10.37

Complaint per force
report

1 of 4

1 of 18

1 of 19

1 of 42

1 of 49

1 of 18

1 of 17

1 of 10

197

244

77

48

39

50

76

600

Complaint/Part I
Crimes (1,000)
Complaint per crime

Total Number
External Discourtesy
Complaints

165

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The focus of this section was to compare external complaints for the use of force and
discourtesy against a set of important departmental workload measures (please see Chapter Six
for a similar set of analyses for the use of force). In doing so, a few cities distinguished
themselves from the rest.
In terms of beneficial outcomes, two departments were ahead of the rest. That is, St.
Petersburg, by far, had the fewest use of force citizen complaints per citizen call for service,
crime, arrest, and force report. With respect to discourtesy citizen complaints, St. Petersburg had
the fewest per arrest and force report, and the second fewest per call and crime. CharlotteMecklenburg also distinguished themselves, as they had the fewest discourtesy citizen
complaints per call and crime, while they had the second fewest discourtesy complaints per
arrest.
On the opposite end of the spectrum, three departments, across the four measures, were
found to exhibit the least beneficial comparisons. That is, Colorado Springs officers had the
most force and discourtesy citizen complaints per use of force report and crime (tied with
Columbus). Portland officers were found to have the most force and discourtesy citizen
complaints per citizen call. Finally, Columbus officers had the most force and discourtesy
citizen complaints per arrest, and tied with Colorado Springs for the most discourtesy citizen
complaints per crime.

166

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Use of Force and Discourtesy Officer Totals
The next analysis extends the examination of citizen complaints for the use of force and
discourtesy to the officer level. That is, we examine, across the eight cities, the distribution of
such complaints across officers, in shedding light on the range of complaint totals (per officer),
mean complaint amounts across officers, and the percentage of sworn personnel with a force
and/or discourtesy citizen complaint. In doing so, we restrict our analyses to citizen complaints
for the use of force and discourtesy where an individual officer (either by name or badge
number) was identified. This decision was based on the fact that we did not want to
misrepresent individual level totals by including unnamed officers. For example, in Columbus
there were 242 complaints made against the “Division of Police” as a whole, and if we included
these allegations, it would appear that a single officer had this many complaints over the twoyear study period. As such, we excluded such cases and our external complaints for the use of
force and discourtesy totals differ from that cited previously in Table 7-1.52
Table 7-9 displays a variety of officer-level totals regarding external complaints for the
use of force and discourtesy. We begin by presenting the number of officers (in our columns)
that received x number of complaints. As we examine the distribution of citizen complaints
across our cities, we find that two-year totals range from one complaint (in every city) to
fourteen (in Columbus). It is apparent, that in some cities (e.g., St. Petersburg and Fort Wayne)
there is a much tighter clustering in terms of the number of complaints per officer (at the lower
end) than that found in others (e.g., Columbus and Portland), where officer totals are more
dispersed (and at higher levels).
52

The following represents the total number of complaints, per city, that did not name an individual officer (and thus
were excluded): Colorado Springs (2), Portland (13), Albuquerque (1), Fort Wayne (0), St. Petersburg (0), Knoxville
(0). Charlotte-Mecklenburg (0), and Columbus (242).

167

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Beneath these totals we also present the total number of citizen complaints for force and
discourtesy (where an officer was named), mean totals that represent the average number of
complaints per officer53, as well standard deviations. It is apparent that there is a degree of
variation across our cities. For example, Fort Wayne, St. Petersburg, Knoxville, and CharlotteMecklenburg averaged approximately one and a quarter complaint, while Columbus and
Portland averaged close to two complaints per officer.
The last three rows in Table 7-9 represent the number of officers that made up the
complaint totals (per city), the total number of sworn personnel in each department (as a
potential pool of officers that could receive a citizen complaint), and the percentage of sworn
personnel that were in the pool of officers with a complaint. For example, 231 officers in
Colorado Springs comprised the 387 total citizen complaints for force and discourtesy, while
roughly 35% of their sworn personnel had at least one of these complaints. Across the cities, we
found that St. Petersburg and Charlotte-Mecklenburg had the fewest percentage of sworn
personnel with a citizen complaint for force and/or discourtesy (9.2 and 10.3, respectively),
while Colorado Springs (34.5%) and Portland (26.7%) were at the other end of the spectrum.

53

Note that these means do not include organizational members that did not receive any complaints. As such, these
figures represent the average number of complaints among those officers that had at least one (over the two year
period).

168

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-9. External Complaints for Force or Discourtesy: Officer Totals
Col. Sp. Port.

Albq.

Ft. Wy. St. Pete Knox.

Char.

Colum.

Number of
Complaints
1
2
3
4
5
6
7
8
9
10
11
12
13
14

146
50
16
7
8
3
1
0
0
0
0
0
0
0

150
60
22
17
7
2
1
2
0
2
1
0
0
0

91
33
7
4
3
1
0
0
0
0
0
0
0
0

63
8
1
0
0
1
0
0
0
0
0
0
0
0

40
7
0
0
1
0
0
0
0
0
0
0
0
0

57
10
2
1
0
0
0
0
0
0
0
0
0
0

136
26
5
0
1
0
0
0
0
0
0
0
0
0

244
101
51
15
11
3
9
3
1
1
1
0
0
1

N Complaints
Mean
S.D.

387
1.68
1.16

505
1.91
1.56

215
1.55
0.97

88
1.21
0.87

59
1.23
0.66

87
1.24
0.58

208
1.24
0.56

863
1.96
1.63

N Officers
Total # Sworn

231
669

264
989

139
986

73
457

48
520

70
382

168
1638

441
1819

Complaint Officers/ 34.53
Sworn

26.69

14.10

15.97

9.23

18.32

10.26

24.24

169

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
The purpose of this section of this chapter was to extend the analyses of citizen
complaints for the use of force and discourtesy to the officer level. In doing so, we presented
(per agency) the number (and averages) of citizen complaints officers received, as well as
comparisons to the overall population of sworn personnel. The results revealed some interesting
comparisons. First, the range of complaints per officer, across sites, varied from one to 14. In
addition, the average total complaints (among officers identified by the citizen) was the lowest
for Fort Wayne, St. Petersburg, Knoxville, and Charlotte-Mecklenburg, while the highest was
found in Columbus and Portland. Second, in examining the percentage of departmental sworn
personnel with at least one citizen complaint, we found variation across cities. The fewest
percentage of sworn personnel with a citizen complaint for force and/or discourtesy was found in
St. Petersburg and Charlotte-Mecklenburg, with the highest in Colorado Springs.

Multivariate Analyses
The final analysis extends the preceding work by examining external use of force and
discourtesy complaints from a multivariate framework. The purpose here is to determine if some
cities have a higher probability of receiving complaints for force and discourtesy, while
controlling for theoretically relevant factors. For this analytical model, we include three
characteristics of the citizen (i.e., sex, race, and age) who filed the complaint. Citizen sex and
gender are binary variables where male complainants and non-white complainants were coded as
one. Citizen age reflects the complainant’s age in years at the time the complaint was filed. In

170

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

addition, we include dummy variables which measure the city where the complaint was filed.54
Descriptive statistics for all variables used in the multivariate analysis is provided in Table 7-10.
Table 7-11 presents the logistic regression results predicting external use of force and
discourtesy complaints. When controlling for citizen characteristics, every city had a
significantly less likelihood of receiving a force or discourtesy complaint when compared to Fort
Wayne. Although all effects were substantial, Albuquerque and St. Petersburg were the cities
with the most pronounced differences. Officers in both of these cities were least likely, than the
rest of the cities, to have force or discourtesy complaints made against them than officers in Fort
Wayne. Finally, we found that none of the citizen characteristics were statistically significant.
Summary
The primary aim of this section was to utilize a multivariate approach to predict use of
force and discourtesy citizen complaints. In doing so, we were able to isolate the independent
effects of each city, controlling for the other cities as well as theoretically relevant citizen
controls (i.e., sex, race, and age). In terms of comparing cities, we found in every instance that
our seven research sites had a significantly lower likelihood of receiving a use of force or
discourtesy citizen complaint than our reference category - Fort Wayne. The most pronounced
differences, from Fort Wayne, were found for Albuquerque and St. Petersburg. In terms of our
citizen controls, none were statistically distinguishable in our multivariate model.

54

A series of multivariate models were run to determine which site would best serve as the reference category for
the final model. Since there were few theoretical reasons to choose one site over the other, multivariate comparisons
were made using each site as the reference category. In the end, the results for Fort Wayne provided the most
consistent results across all models. That is, in nearly every case, Fort Wayne was found to have significantly more
complaints than each of the other seven sites. Given this information, as well as the extremely high percentage of
external use of force and discourtesy complaints (per all complaints) noted in Table 7-1, we use Fort Wayne as the
reference category in our analytical model.

171

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-10. Descriptive Statistics for Multivariate Model
Variable

Range

Mean

Std. Dev.

Dependent
Use of Force Complaint

0-1

.20

.40

Discourtesy Complaint

0-1

.20

.40

Combined Complaint

0-1

.41

.49

Citizen Sex

0-1

.56

.49

Citizen Race

0-1

.54

.50

11-79

36.2

12.49

Colorado Springs

0-1

.13

.34

Portland

0-1

.20

.40

Albuquerque

0-1

.15

.36

Fort Wayne

0-1

.01

.18

St. Petersburg

0-1

.04

.19

Knoxville

0-1

.02

.15

Charlotte-Mecklenburg

0-1

.06

.24

Columbus

0-1

.38

.49

Independent

Citizen Age

172

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 7-11. Logistic Regression Predicting Citizen Complaints
Reference Category: Fort Wayne
Variable

B

S.E.

Sig.

Exp(B)

Citizen Sex

.09

.07

.170

1.10

Citizen Race

.00

.07

.956

1.00

Citizen Age

.00

.00

.223

1.00

Colorado Springs

-2.74

.43

.000

.06

Portland

-3.20

.43

.000

.04

Albuquerque

-3.86

.44

.000

.02

St. Petersburg

-3.84

.45

.000

.02

Knoxville

-3.10

.49

.000

.05

Charlotte-Mecklenburg

-2.50

.45

.000

.08

Columbus

-2.85

.43

.000

.06

Constant

2.74

.44

.000

15.53

Chi Square
- 2 Log Likelihood
Pseudo R square
N=4,060

262.10
5239.32
.084

173

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 8
Civil Litigation Data

The final piece of the second phase of the project involved the collection and analysis of
civil litigation data. In this chapter, we examine these data to investigate how often citizens
alleged police misconduct (particularly the misuse of force), the outcome or dispositions of such
allegations, and the monetary payout cities have made.

Methodology
Similar to use of force and complaint data, at each of the sites we collected 24 months of
data within the same timeframe as noted in Chapters 6 and 7. By far, attempting to gather civil
litigation data proved more challenging than any other aspect of the project; and even in those
cities where we were successful the substance of the information collected was often limited.
Many of the difficulties we experienced have been discussed by previous scholars (Cheh, 1995,
Fyfe, 1986, Worrell, 2001), but perhaps best summarized by Archbold and Maguire (2002) with
respect to insufficient departmental data collection protocols and/or a department or city’s
reluctance to release such information to researchers.
In Albuquerque and Columbus the city attorney to the police department chose not to
make these data available to our research team. In Albuquerque we were continually promised
that the data would be disclosed without success. A somewhat similar scenario played out in
Columbus.55 Hence, Albuquerque and Columbus failed to submit the requested data. We leave
it to the reader to draw conclusions as to why this may be the case, and the implications failure to
55

The city attorney for the police department in Columbus initially released cursory information on lawsuits filed,
but failed to respond to any future requests detailing the specifics of these lawsuits; in Albuquerque the city attorney
for the police failed to release any information at any time.

174

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

disclose the data may have on the findings. In Charlotte-Mecklenburg, police department
officials were willing to collect and disclose the data, but in the end we were unable to obtain
and analyze it prior to the conclusion of the project. Finally, Knoxville provided us with civil
litigation data, but the format was insufficient for our purposes so as to draw any substantive
conclusions.
The remaining four cities were willing and able to disclose civil litigation data in a
format that we could use to compare cities. We caution, however, that the information provided
is still limited in scope. Specifically, the only information we were fully and confidently able to
glean included the number of lawsuits filed during our two-year study period, the outcome or
dispositions of these cases, and the extent of monetary payout. As a result, these data are what
we were able to rely on for the following analyses.

Analyses and Findings
Given the relatively few civil litigation cases filed (in comparison with force or
complaints), we do not place the number of lawsuits within the context of calls for service,
reported crimes, or arrests. Rather, we examine the findings as illustrated in Table 8-1.

175

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 8-1. Lawsuits Filed, Outcomes, and Monetary Payouts
Col. Sp.

Port.

Ft. Wy.

St. Pete

# All
Lawsuits

17

75

91

22

# Force
Lawsuits

3
(17.6%)

48
(64.0%)

51
(56.0%)

3
(13.6%)

All
Lawsuits
Outcome

Dismissed=10
Pending=3
Settled=4

Dismissed=15
Pending=27
Arbitration-win=1
Arbitration-loss=1
Settled=31

Dismissed=7
Pending=11
Denied=43
Settled=22
Loss=8

Dismissed=4
Pending=6
Settled=11
Loss=1

Force
Lawsuits
Outcome

Dismissed=2
Pending=1

Dismissed=7
Pending=20
Arbitration-win=1
Arbitration-loss=1
Settled=19

Dismissed=3
Pending=7
Denied=21
Settled=14
Loss=6

Dismissed=1
Settled=2

All
Lawsuits
Monetary
Payout

$5,259,32656

$1,292,649

$113,89957

$145,89958

Force
Lawsuits
Monetary
Payout

$0

$967,656

$54,67459

$5,500

56

$5,250,000 of this amount was paid as part of one settlement.

57

This figure computed based on 20 of the 30 settled/loss cases. Monetary figures were not reported for the
remaining 10 cases.
58

This figure computed based on the 11 settled cases. The one case the city lost was being appealed.

59

This figure computed based on 11 of the 20 settled/loss cases. Monetary figures were not reported for the
remaining 9 cases.

176

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The top two rows indicate both the total number of lawsuits filed during the two-year
study period as well as the number and percentage of the total that allege some form of improper
force usage. As shown, there is a fairly clear distinction between the two cities with the most
lawsuits filed (i.e., Fort Wayne and Portland) and the two cities with the least lawsuits filed (i.e.,
St. Petersburg and Colorado Springs). Fort Wayne experienced the most suits filed overall
(n=91), as well as for force (n=51). Portland had the second highest frequency of lawsuits filed
overall (n=75), as well as for force (n=48). These two cities also had a relatively high
percentage of their overall lawsuits involve allegations of improper force (56.0% in Fort Wayne
and 64.0% in Portland). St. Petersburg and Colorado Springs were at the other end of the
spectrum. St. Petersburg received 22 lawsuits with three (13.6%) being for force, while
Colorado Springs received 17 lawsuits with three (17.6%) being for force.
The middle two rows in Table 8-1 indicate the case dispositions of all the lawsuits filed
as well as the case dispositions for those cases involving force. As illustrated, while there are
some similarities as to how the cities label various types of disposition outcomes (i.e., dismissed,
pending, settled, loss), there are also some unique categorizations within some cities. In
particular, Portland had two cases “arbitrated” (one for a win and one for a loss), while Fort
Wayne had a substantial number of claims (n=43) “denied” (i.e., cases that failed to meet
minimal evidence of wrongdoing to move forward in the civil system).
Perhaps the best way to glean meaningful substance from these outcomes is to compute
the percentage of cases where the city either lost or settled the case in relation to the total
number of lawsuits filed. Using this strategy and beginning with the all lawsuits filed row, St.
Petersburg lost/settled the highest percentage of cases at 54.5 percent (i.e., 12 of the 22 cases).

177

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

This is followed by Portland, which settled/lost 32 of their 75 cases for a percentage of 42.7.
Fort Wayne had the next highest percentage at 33.0 (30 of 91 cases). Finally, Colorado Springs
fared the best with losing or settling just 23.5 percent of the time (4 of 17 cases).
Applying a similar strategy for force dispositions (i.e., row 4), we see that St. Petersburg
lost/settled the highest percentage of cases at 66. percent (i.e., 2 of the 3 force cases). This is
followed by Portland, which settled/lost 20 of their 48 cases for a percentage of 41.7. Fort
Wayne had the next highest percentage at 39.2 (20 of 51 cases). Finally, Colorado Springs fared
the best losing or settling none of their three force cases.
Finally, the last two rows of Table 8-1 depicts the monetary payout the cities made in
regard to their lawsuits during the two-year study period. As shown, Colorado Springs had the
highest payout of the four cities at $5,259,326. However, it is important to note that nearly all of
this figure ($5,250,000) resulted from one case involving a number of officers over a dispute
involving the city pay structure, including overtime. Moreover, none of the cases involving
force involved a monetary payout. Portland had the next highest payout at $1,292,649 for all
lawsuits and $967,656 for lawsuits involving force. The remaining two cities had substantially
less payout. St. Petersburg ended up paying $145,899, but only $5,500 of the total amount was
for force cases. Fort Wayne paid out a little less overall for their lost/settled cases at $113,899,
but more in terms of force cases ($54,674). A degree of caution should be noted when
interpreting the monetary payouts as indicated in the footnotes at the bottom of Table 8-1 - as the
figures are reported based on the available information.

178

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
Although we were only able to collect and analyze a limited amount of information
pertaining to civil litigation across four of our cities several important findings emerge.
Colorado Springs faired extremely well. Over the course of the two-year study period, only
three lawsuits pertaining to the misuse of force were filed. Of these, two were dismissed and one
was still pending at the conclusion of the project. Viewed alternatively, Colorado Springs had
no misuse of force cases settled or lost during the study and subsequently paid out no money.
Not far behind Colorado Springs was St. Petersburg which settled just two cases. Moreover, the
payout for these two cases cost just $5,500 in total.
Conversely, both Portland and Fort Wayne had 20 cases that involved the improper use
of force that were either settled or lost during the study period. However, the monetary payout
between the two cities was substantial as Portland paid nearly one million dollars for these
lawsuits, while Fort Wayne ended up paying just over $50,000. One has to be cognizant though
to the fact that monetary figures were not reported in nine of the 20 Fort Wayne cases, which
could increase their payout significantly and bring it more on par with Portland.

179

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 9
Project Summary Findings

Phase I
In Chapter 3 we presented results of a national mail survey designed to assess the various
types of written use of force policies police agencies use throughout the country, particularly
those that incorporate the use of a force continuum. What follows is a summary of the key
findings:
Force Continuum Policies
<

Over 80 percent of the responding agencies use some type of force continuum
policy. Of these agencies, the linear design was the most frequently used (73%),
followed by matrix/box designs and circular/wheel designs, each with roughly 10
percent of the agencies using them.

<

With respect to the tactical placement of force tactics (soft hands, pain
compliance controls, hard hands) and weapons (batons, chemical sprays, CEDs),
and how police agencies rank the order of such in terms of progression, the key
finding uncovered was the enormous variation present.

<

<

Of the 476 agencies that outlined their force progression, a total of 123
different permutations were uncovered, ranging from three to nine
different levels (the vast majority, 390 or 82.0%, relied on five or six
levels).

<

Of the 371 agencies that listed citizen resistance progression, a total of 23
permutations were uncovered, ranging from three to seven different levels
(the vast majority, 343 or 92.4%, relied on five or six levels).

For departments that attempt to incorporate a force continuum approach into their
policy and provide officers with explicit guidance as to the types of force most
appropriate given varying types of resistance (i.e., link citizen resistance to force),
it was difficult to identify a typically used preference.
<

The most frequent approach was used by only 20 percent of the
departments, while the second and third most frequently used approaches
were just half of that at 10 percent.

180

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

The placement of chemical sprays and CEDs appear to offer the greatest
challenge for police administrators as to the proper placement within the force
continuum.
<

Roughly 30 percent of the agencies place chemical sprays with pain
compliance techniques, another 30 percent of the departments place
chemical sprays with hard hand tactics, and just over a third of the
agencies place chemical sprays with impact weapons. A handful of
departments even place chemical sprays as low as empty-hand soft
techniques (n=12) and as high as deadly force (n=6).

<

Compared to chemical sprays, there was somewhat less variation when it
comes to CED placement, but far from a clear consensus. Nearly 60
percent of the agencies place CEDs at the impact weapon level, while
another 2.0 percent place it along with deadly force. The remaining
agencies place CEDs with some sort of hands on force (a quarter of the
agencies place CEDs at the same level as hard empty-hand tactics, with
another 13.1 percent placing CEDs with pain compliance techniques).

Phase II
In the second phase of the project we selected eight police agencies for in-depth study on
use of force policies and practices. Each agency had a different policy approach with respect to
instructing officers on the application of force, and we sought to determine which types of
policies offer more beneficial outcomes to police practitioners. To do so, we relied on four
primary sources of data: a patrol officer survey, use of force report data, complaint data, and
civil litigation data. Here we present the key findings gleaned from each of these sources of
data.

Patrol Officer Survey Data
In Chapter 5, our patrol officer survey, we captured information on perceptions of the
overall less lethal use of force policy, relevant outcomes of the policy, as well as the training

181

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

associated with the policy. What follows is a summary of key findings of each of these primary
sections.
Less Lethal Force Policy Perceptions
In assessing officers’ perceptions of their less lethal use of force policy, we utilized
responses from survey items that focused on policy assistance, restrictiveness, clarity, guidance,
and review fairness. In doing so, we rank order cities based on their overall agreement and
disagreement (i.e., combining “strongly” and “somewhat” responses) with each of the relevant
survey items. At the end of this section, we present cumulative rank ordering across the
following five major policy perception areas. For all rankings in this section, lower scores
indicate more positive or favorable perceptions.
Assistance
We begin by ranking each department in terms of their agreement with the survey
question that their less lethal force policy “assists officer decision making.” In terms of
assistance, Knoxville officers expressed the most positive assessments, while St. Petersburg
expressed the least favorable evaluations.
<

Rankings:
1-Knoxville, 2-Albuquerque, 3-Fort Wayne, 4-Columbus,
5-Colorado Springs, 6-Portland, 7-Charlotte-Mecklenburg, 8-St. Petersburg.

Restrictiveness
In a similar manner, we also rank each department in terms of their disagreement with the
survey question that their less lethal force policy “is too restrictive.” St. Petersburg officers
expressed the strongest beliefs that their policy was restrictive, while Knoxville officers were at
the other end of the spectrum (i.e., policy is not too restrictive).
182

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

Rankings:
1-Knoxville, 2-Colorado Springs, 3-Fort Wayne, 4-Albuquerque,
5-Columbus, 6-Portland, 7-Charlotte-Mecklenburg, 8-St. Petersburg.

Clarity
Next, we present the rank ordering of departments in terms officers’ agreement that their
less lethal use of force policy “is clear.” The most favorable assessments of policy clarity come
from Knoxville, while Portland officers held the least positive perceptions that their policy is
clear.
<

Rankings:
1-Knoxville, 2-Albuquerque, 3-Charlotte-Mecklenburg, 4-Fort Wayne,
5-Colorado Springs, 6-Columbus, 7-St. Petersburg, 8-Portland.

Guidance
In ranking departments in terms of guidance, we utilized agreement levels across the
survey item that asked officers the extent to which their policy provides guidance in terms of
“when force can and cannot be used.” Across departments, Portland ranked last in their
agreement that the policy provides them with adequate guidance. On the other hand, Knoxville,
as they did for assistance and clarity, held the strongest positive attitudes that their policy
adequately guides their use of force.
<

Rankings:
1-Knoxville, 2-Albuquerque, 3-Charlotte-Mecklenburg, 4-Columbus,
5-St. Petersburg, 6-Colorado Springs, 7-Fort Wayne, 8-Portland.

183

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Review Fairness
Our final topic, in assessing overall less lethal force policy perceptions, departs from the
previous four areas that focused on policy interpretation and utilization, and instead deals with
the review process after use of force is used. Below, we present the departmental ranking of
assessments that the “supervisory review of use of force reports is fair.” The most positive
attitudes in supervisory review fairness were found among Portland officers, while the least
positive attitudes were found in Colorado Springs.
<

Rankings:
1-Portland, 2-Knoxville, 3-Columbus, 4-Fort Wayne,
5-Charlotte-Mecklenburg, 6-Albuquerque, 7-St. Petersburg, 8-Colorado Springs.

Summary
While this section illuminated differences across departments in ranking their perceptions
of core areas of their use of force policy individually, one might also wonder how departments
ordered in terms of these dimensions cumulatively. Table 9-1 does just that by combining the
ranking results from the previous five sub-sections to give a more global characterization of
departmental ranking of their less lethal use of force policy in general. Not surprisingly,
Knoxville, leads all departments, by far, in their overall assessment of their force policy, while
St. Petersburg is last.
<

Rankings:
1-Knoxville, 2-Albuquerque, 3-Fort Wayne, 4-Columbus,5-CharlotteMecklenburg, 6-Colorado Springs, 7-Portland, 8-St. Petersburg.

184

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-1. Perceptions of LTLF Policy Comparison Rank: Cumulative Results.

Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Assistance
Rank

5

6

2

3

8

1

7

4

Restrictiveness
Rank

2

6

4

3

8

1

7

5

Clarity
Rank

5

8

2

4

7

1

3

6

Guidance
Rank

6

8

2

7

5

1

3

4

Supervisory
Review Rank

8

1

6

4

7

2

5

3

Total

26

29

16

21

35

6

25

22

Overall Rank

6

7

2

3

8

1

5

4

*Lower ranks indicate more favorable perceptions of the policy

185

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Outcomes of Less Lethal Force Policy
In addition to evaluating agencies in terms of the use of force policy in general, we were
also interested in assessing views of the policy in terms of important outcomes, all of which have
been studied behaviorally as part of this project. That is, we asked officers what impact, in their
estimation, the less lethal use of force policy had on potential citizen injuries (studied
behaviorally in Chapter 6), officer injuries (studied behaviorally in Chapter 6), citizen
complaints (studied behaviorally in Chapter 7), and lawsuits (studied behaviorally in Chapter 8).
Officers were asked to respond to these questions with one of the following categories:
“increases the likelihood,” “decreases the likelihood,” or “neither increases nor decreases the
likelihood.”
Potential Citizen Injuries
In terms of citizen injuries, the modal response by patrol officers from all departments
was that their agency’s force policy decreases the likelihood of potential citizen injuries.
Variation in these modal responses allow us to rank order agencies with lower values indicating
more positive assessments (i.e., highest percentage of respondents reporting that the policy
decreases the likelihood). As was the case for overall views of the force policy, Knoxville
ranked first. On the other end of the spectrum was Fort Wayne, which had the fewest percentage
of respondents that reported that the policy decreased the likelihood for potential citizen injuries.
<

Rankings:
1-Knoxville, 2-St. Petersburg, 3-Albuquerque, 4-Colorado Springs,
5-Charlotte-Mecklenburg, 6-Portland, 7-Columbus, 8-Fort Wayne.

186

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Potential Officer Injuries
When asked what the effect of the force policy has on potential officer injuries, six of the
agencies’ modal responses was “decreases the likelihood,” while St. Petersburg and CharlotteMecklenburg were divergent, and both had a modal response that their policy actually increases
officer injury potential. Given that the latter would suggest negative assessments (more so than
the lowest percentage of responses found for “decreases the likelihood”), we still rank order
departments from most to least positive. Once again, Knoxville held the most favorable policy
attitudes, while St. Petersburg held the least positive assessments of their policy (i.e., in terms of
the likelihood of officer injuries).
<

Rankings:
1-Knoxville, 2-Colorado Springs, 3-Albuquerque, 4-Columbus,
5-Portland, 6-Fort Wayne, 7-Charlotte-Mecklenburg, 8-St. Petersburg.

Potential Citizen Complaints
Assessing attitudes regarding force policy and citizen complaints was more problematic
than the previous two outcomes. While approximately a third of all respondents, across all
departments, indicated that the policy increased the likelihood of citizen complaints, the modal
response in six of our eight departments was that officers believed that their policy neither
increases nor decreases the likelihood of potential citizen complaints, while Colorado Springs
and St. Petersburg had modal responses which indicated that officers believed that the policy
increases the potential for citizen complaints. Because responses of “neither increases nor
decreases the likelihood” suggest no impact, departments were unable to be scored against one
another.

187

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Potential Lawsuits
Like that for the outcome of citizen complaints, views of the force policy’s effect on
potential lawsuits, with a stronger intensity, were generally negligible across most departments
(and thus not ranked). The modal response in six of our eight department was that officers
believed that their policy neither increases nor decreases the likelihood of potential lawsuits,
while St. Petersburg and Knoxville’s modal response indicated that officers believed that the
policy decreases lawsuit potential.
Summary
In shedding light on officers’ views of the potential impact of the force policy on a
variety of outcomes, modal responses suggest that rank ordering agencies was less problematic
for assessing citizen and officer injuries. Knoxville held the most favorable assessments of the
policy’s effect on both citizen and officer injuries. The least positive evaluations of the policy
were found for Fort Wayne (for citizen injuries) and St. Petersburg (for officer injuries). On the
other hand, beliefs regarding citizen complaints and lawsuits, for most departments, were
negligible.
Training and Less Lethal Force Policy
Our final assessment of officer perceptions of less lethal force has to do with the training
that police receive, which presumably incorporates aspects of the agency’s policy. While the
majority of patrol officers, from all departments, agreed that their training was adequate, we
found substantial variation across sites. What follows is a ranking of departments for pre-service
and in-service training separately, as well as cumulatively (in adding up rank totals across the
two dimensions).

188

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Pre-Service Training
In terms of academy (or pre-service) training on less lethal force, the most favorable
assessments were found in Knoxville (see Table 9-2). On the other end of the spectrum were St.
Petersburg officers, who reported the least positive views of pre-service force training.
<

Rankings:
1-Knoxville, 2-Albuquerque, 3-Columbus, 4-Charlotte-Mecklenburg,
5-Fort Wayne, 6-Portland, 7-Colorado Springs, 8-St. Petersburg.

In-Service Training
With respect to in-service force training, we also found substantial variation across
departments, and also generally less positive attitudes than those noted for pre-service training.
The most favorable views of in-service force training were found in Columbus, while the least
favorable were among Fort Wayne patrol officers.
<

Rankings:
1-Columbus, 2-Knoxville, 3-Charlotte-Mecklenburg, 4-Albuquerque,
5-Portland, 6-St. Petersburg, 7-Colorado Springs, 8-Fort Wayne.

Summary
The last row in Table 9-2 reports the overall ranking by combining departmental
positions noted for pre-service and in-service training. As was the case for overall policy
perceptions, Knoxville led the rest of the departments in their positive assessments of training.
The least positive views of overall force training was found in Colorado Springs and St.
Petersburg.
<

Rankings:
1-Knoxville, 2-Columbus, 3-Albuquerque, 4-Charlotte-Mecklenburg,
5-Portland, 6-Fort Wayne, 7-Colorado Springs, 7-St. Petersburg.
189

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-2. Perceptions of LTLF Pre-Service and In-Service Training Comparison Rank
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

LTLF PreService Rank

7

6

2

5

8

1

4

3

LTLF InService Rank

7

5

4

8

6

2

3

1

Total

14

11

6

13

14

3

7

4

Overall Rank

7

5

3

6

7

1

4

2

*Lower ranks indicate more positive assessments

190

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
Finally, in an effort to summarize agency rankings further, we assessed the combined
rankings of officers’ perceptions of the overall policy (from Table 9-1), the policy impact on
potential citizen and officer injuries (see pages 185-186), and pre-service and in-service training
associated with the policy (from Table 9-2). These three rankings were added together to come
up with an overall assessment across cities. As such, the city with the most favorable
perceptions, across the board, was Knoxville, while the least favorable cumulative assessments
were found in St. Petersburg and, to a lesser extent, Portland and Fort Wayne.
Overall Policy
Ranking
1-Knoxville
2-Albuquerque
3-Fort Wayne
4-Columbus
5-Charlotte-Mecklenburg
6-Colorado Springs
7-Portland
8-St. Petersburg

Citizen/Officer Injury
Ranking
1-Knoxville
2-Colorado Springs
2-Albuquerque
4-St. Petersburg
5-Portland
5-Columbus
7-Charlotte-Mecklenburg
8-Fort Wayne

Training
Ranking
1-Knoxville
2-Columbus
3-Albuquerque
4-Charlotte-Mecklenburg
5-Portland
6-Fort Wayne
7-Colorado Springs
7-St. Petersburg

Combined Policy Perceptions
Ranking
1-Knoxville
2-Albuquerque
3-Columbus
4-Colorado Springs
5-Charlotte-Mecklenburg
6-Portland
6-Fort Wayne
8-St. Petersburg

191

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Force Report Data
In Chapter 6, we looked at a multitude of ways to compare cities in relation to how
officers use force, the nature of resistance officers face, and the combination of the two. In
addition, we examined suspect and officer injuries. Here we attempt to summarize key findings,
and more importantly, place the findings into a meaningful context so readers can more easily
assess the merits/drawbacks of the various policy approaches.
Aggregate Officer Force Usage in Relation to Calls for Service, Reported Crimes,
and Arrests
Table 9-3 ranks each of the cities with respect to how their force usage varies when
placed within the context of calls for service, reported crimes, and arrests.60 Depicted across
each of the rows is where the cities ranked relative to the indicator (i.e., calls, crimes, arrests).
Lower numbers indicate less force, while higher numbers indicate more force. These ranks are
then summed up for a composite ranking across all three indicators. As shown, Colorado
Springs faired the best (i.e., officers used the least amount of force), while Fort Wayne fell on the
opposite end of the spectrum.(i.e., officers used the most amount of force).
<

Rankings:
1-Colorado Springs, 2-Charlotte-Mecklenburg, 3-Knoxville, 4-Albuquerque, 5-St.
Petersburg, 6-Columbus, 7-Portland, 8-Fort Wayne.

60

The summary tables depicted here draw on more detailed tables that were presented in Chapter 6.

192

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-3. Force Comparison Rank for Calls for Service, Reported Crimes, and Arrests
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Calls
Crimes
Arrests

1
2
1

8
7
6

3
3
4

7
8
7

6
5
5

2
4
3

4
6
8

5
1
2

Total

4

21

10

22

16

9

18

8

Rank

1

7

4

8

5

3

6

2

Table 9-4. Force Comparison Rank to by Force Type
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Weaponless
Soft Hands
Hard Hands

1
7

4
3

2
5

6
8

5
2

7
6

3
1

8
4

Weapon
Chemical
Baton
CED
Impact
Point
Deadly
Other

5
2
8
1
3
7
4

3
4
7
8
8
6
7

8
8
5
7
4
1
6

7
6
1
6
1
4
2

4
1
6
1
5
5
3

2
6
4
4
5
1
5

6
3
2
1
2
1
1

1
5
3
5
7
8
8

Total

38

50

46

41

32

40

20

49

Rank

3

8

6

5

2

4

1

7

*Lower ranks indicate less force, higher ranks indicated more force

193

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Aggregate Officer Force
Table 9-4 incorporates a similar ranking system to Table 9-3 for tallying and ranking
force, but in this instance examines the amount of force used by various types. Examining force
in this manner shows that Columbus officers, by far, used the least amount of force. Conversely,
Portland officers used the most force overall.
<

Rankings:
1-Columbus, 2-St. Petersburg, 3-Colorado Springs, 4-Knoxville,
5-Fort Wayne, 6-Albuquerque, 7-Charlotte-Mecklenburg, 8-Portland.

Aggregate Suspect Resistance
Table 9-5 follows the same approach by ranking the amount of suspect resistance officers
faced by type. The difference here is that the rankings are opposite. In these cases, lower
numbers indicate more resistance, while higher numbers indicate less resistance. Hence,
Knoxville officers faced the most challenging suspects (i.e., those who resist the most), while
Fort Wayne officers faced the least challenging suspects (i.e., those who resist the least).
<

Rankings:
1-Knoxville, 2-Charlotte-Mecklenburg, 3-Columbus, 4-Albuquerque,
5-Colorado Springs, 6-St. Petersburg, 7-Portland, 8-Fort Wayne.

194

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-5. Resistance Comparison Rank for Resistance Types
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Fail to Comply
Defensive
Aggressive
Deadly

6
3
6
7

2
8
5
8

8
7
2
2

7
6
7
5

4
4
8
6

5
2
1
1

1
5
4
3

3
1
3
4

Total

22

23

19

25

22

9

13

11

Rank*

5

7

4

8

5

1

3

2

*Lower ranks indicate more resistance, higher ranks indicated less resistance

195

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Force Relative to Resistance
The primary reason for reversing the ranking system between Tables 9-4 and 9-5 is to
compare “relative” force to resistance as depicted in Table 9-6. For instance, a city ranking 8th
on force and 8th on resistance (summing to 16) indicates a city where officers use a lot of force,
but face little resistance, relatively. Conversely, a city ranking 1st on force and 1st on resistance
(summing to 2) indicates a city where officers use little force, but face a lot of resistance,
relatively. Those cities that have a lot of force and resistance, or little force and resistance are
treated equally. For example, a city ranking 8th on force and 1st on resistance (i.e., a lot of force
and resistance) would sum to 9, and rank identical to a city ranking 1st on force and 8th resistance
(i.e., a little force and resistance). Thus, a higher number and rank indicates a city where officers
are using relatively more force than the resistance faced, while a lower number and rank
indicates a city where officers face relatively more resistance than the force used. Within this
context, Table 9-6 shows that Columbus officers used the least amount of force relative to
resistance faced. Conversely, Portland officers used a relatively high amount of force compared
to the level of resistance faced.
<

Rankings:
1-Columbus, 2-Knoxville, 3-St. Petersburg, 4-Colorado Springs,
5-Charlotte-Mecklenburg, 6-Albuquerque, 7-Fort Wayne, 8-Portland.

196

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-6. Force Relative to Resistance Comparison Rank
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Force
Rank

3

8

6

5

2

4

1

7

Resistance
Rank

5

7

4

8

5

1

3

2

Total

8

15

10

13

7

5

4

9

Overall Rank

4

8

6

7

3

2

1

5

Table 9-7. Force Ranking Results from Table 9-3 and Table 9-6
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Colum.

Char.

Table 9-3
Rank

1

7

4

8

5

3

6

2

Table 9-6
Rank

4

8

6

7

3

2

1

5

Total

5

15

10

15

8

5

7

7

Overall Rank

1

7

6

7

5

1

3

3

*Lower ranks indicate less force, higher ranks indicated more force

197

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
Taking the aggregate levels findings from Table 9-3 and incorporating them with the
findings presented in Table 9-6 allows us to make a more definitive assessment as to which cities
fare better in terms of how frequently they use force, while also accounting for the nature of
suspect resistance. As Table 9-7 shows, when summing and comparing the rankings, Colorado
Springs and Knoxville fare relatively well; Charlotte-Mecklenburg, Columbus, St. Petersburg,
and Albuquerque fall in the middle range; and Portland and Fort Wayne fare relatively poorly.
<

Rankings:
1-Colorado Springs, 1-Knoxville, 3-Charlotte-Mecklenburg, 3-Columbus,
5-St. Petersburg, 6-Albuquerque, 7-Fort Wayne, 7-Portland.

Force and Resistance Comparisons
After looking at force and resistance in an aggregate manner, we turned to bivariate and
multivariate examinations. In particular, we looked at how officers responded (i.e, the types of
force they used) to specific types of resistance (i.e., failure to comply, defensive physical,
aggressive physical). In addition, we looked at the highest level of force officers use across the
cities controlling for alternative factors than may prompt force. The key findings are as follows:
<

Colorado Springs, Columbus, and St. Petersburg officers used various forms of
hard hand tactics, chemical sprays, and CEDs on suspects who displayed just
failure to comply resistance, despite their policy advising against such force.
<

<

Meanwhile, Albuquerque’ policy permitted the use of CEDs on verbally
resistant suspects but rarely did officers use it.

Portland’s policy advises officers against the use of hard hand tactics on
defensively resistant suspects, while Albuquerque, Knoxville, and CharlotteMecklenburg’s policy permits such force. Yet, officers is all four cities used this
form of force similarly.

198

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

Moreover, Colorado Springs policy, similar to Portland, advises against
hard hand tactics on defensively resistant suspect, yet Colorado Springs
officers used such force more frequently than any of the other cities.

<

St. Petersburg and Charlotte-Mecklenburg generally restrict use of the CED to
instances where suspects are aggressively resistant, yet officers in these two cities
used this type of force much differently despite the similarity in policy (i.e., St.
Petersburg officers used the CED frequently, while Charlotte-Mecklenburg used
the CED infrequently).

<

Consistent with our bivariate analyses, the multivariate analyses indicated that it
is difficult to pinpoint any sort of consistent policy effects. In some cases we
found officers operating along the lines that one might expect given the varying
policy approaches, while in other instances we do not.

Summary
From this set of analyses, we were unable to find a distinct link in terms of officer
behavior operating within the realm of policy - in fact, we often found officers using force on
suspects that appeared to fall outside the policy. Such behavior was not restricted to a certain
city or policy approach. While the findings call into question the connection between policy and
behavior, the reader should interpret this with a degree of caution. First, there are certainly
instances, as illustrated in Chapter 6, where officer behavior falls within the realm of their policy
mandate. Second, mainly of the were bivariate in nature and do not permit control of other rival
causal factors that may explain the reasons officers used the force they did, and thus justify that
the behavior was actually within policy rather than what appears outside of policy (see also
footnotes 43 and 44).

Suspect Injuries and Force
Drawing on data derived from the force reports we also examined suspect injuries. More
specifically, we computed suspect injury rates with respect to the frequency of injuries in

199

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

relation to the total number of force reports filed per city. The following ranks how cities fared
from the city with the least suspect injuries (i.e., St. Petersburg) to the city with the most suspect
injuries (i.e., Charlotte-Mecklenburg).
<

Rankings:
1-St. Petersburg, 2-Columbus, 3-Albuquerque, 4-Fort Wayne,
5-Portland, 6-Colorado Springs, 7-Knoxville, 8-Charlotte-Mecklenburg.

From a policy perspective, we note several interesting findings:
<

The range of injuries across cities varies widely: from 15.9 percent in St.
Petersburg to 73.5 percent in Charlotte-Mecklenburg.

<

Charlotte-Mecklenburg and St. Petersburg both restrict CED use to aggressively
resistant suspects, yet they fall on opposite ends in terms of suspect injuries
(Charlotte-Mecklenburg officers had the most injuries, St. Petersburg the least).
<

Moreover, Colorado Springs officers used a CED far more frequently than
officers in other cities, yet they had just the third highest percentage of
injuries.

<

Arguably, the three cities with the least restrictive policies (Fort Wayne and
Columbus do not link resistance to force, and Albuquerque permits all types of
force on verbally resistant suspects) all fall in the middle to low range on suspect
injuries.

<

Knoxville and Colorado Springs resulted in the second and third highest
percentage of injuries. However, their policies are substantially different.
Colorado Springs employs a wheel design and reserves hard hand tactics and
impact force to aggressive resistant suspects, while Knoxville allows officers to
use not only hard hands tactics and impact methods on defensively resistant
suspects, but also chemical spray and a CED.

Officer Injuries and Force
Drawing on data derived from the force reports we also examined officer injuries. More
specifically, we computed officer injury rates with respect to the frequency of injuries in relation
to the total number of force reports filed per city. The following ranks how cities fared from the

200

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

city with the least officer injuries (i.e., Columbus) to the city with the most officer injuries (i.e.,
Knoxville).61
<

Rankings:
1-Columbus, 2-Portland, 3-St. Petersburg, 4-Fort Wayne,
5-Colorado Springs, 6-Charlotte-Mecklenburg, 7-Knoxville.

From a policy perspective, we note several interesting findings:
<

There is little variation across cities with respect to officer injuries when
compared to suspect injuries.
<

<

Officer injury rates are much lower on average than that of suspect injuries.
<

<

The city with the highest percentage of officer injuries (Knoxville,
14.8%), was still lower than the city with the lowest percentage of suspect
injuries (St. Petersburg, 15.9%).

Charlotte-Mecklenburg and St. Petersburg both restrict CED use to aggressive
resistance, yet Charlotte-Mecklenburg reported the second highest percentage of
officer injuries, while St. Petersburg reported just the fifth highest percentage.
<

61

The small range of injuries goes from 8.1 percent in Columbus up to just
14.8 percent in Knoxville.

Moreover, Colorado Springs officers used a CED far more frequently than
officers in all the other cities, yet this city still had just the third highest
percentage of officer injuries.

<

Two of the cities with the least restrictive policies (Fort Wayne and Columbus)
fall in the middle to low range on officer injuries. One may posit that less policy
restriction may lead to more injuries (i.e., because officers will have more
freedom), but we find that such policies were related to less injuries.

<

Knoxville and Charlotte-Mecklenburg had the highest percentage of injuries, yet
their policies are substantially different. In particular, officers in CharlotteMecklenburg are restricted from using a CED or impact force on suspects unless
they are aggressively resistant. Knoxville’s policy is less restrictive with the use
of a CED and impact weapons, permitting such force to be used on suspects who
display just defensive resistance.

Recall that Albuquerque did not report officer injuries.

201

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary
Finally, in an attempt to summarize agency rankings further we assessed the frequency of
force usage (while also accounting for the nature of suspect resistance - see page 197) as
displayed in column one below, along with a combined ranking of how agencies fared in terms
of suspect injuries (see page 199) and officer injuries (see page 200), as displayed in column two
below, to come up with an overall ranking of how cities compare as displayed in column three
below. Hence, using these multiple indicators gleaned from the use of force report data, we
found that Columbus fared the best in their application of force.
Force Usage
Ranking

Suspect/Officer Injury
Ranking62

Combined Force and Injury
Ranking63

1-Colorado Springs
1-Knoxville
3-Charlotte-Mecklenburg
3-Columbus
5-St. Petersburg
6-Albuquerque
7-Fort Wayne
7-Portland

1-Columbus
2-St. Petersburg
3-Albuquerque
3-Portland
5-Fort Wayne
6-Colorado Springs
7-Charlotte-Mecklenburg
7-Knoxville

1-Columbus
2-Colorado Springs
2-St. Petersburg
4-Knoxville
5-Albuquerque
6-Charlotte-Mecklenburg
6-Portland
8-Fort Wayne

62

For this ranking order we simply summed the ranking of suspect injuries with that of officer injuries. Given that
officer injury data were not captured on the Albuquerque use of force report form, we used an average ranking of 4
(i.e., the middle of the range of 1-8) for officer injuries and then summed to their ranking for suspect injuries.
63

For this ranking order we simply summed the ranking in the force usage column with that of the suspect/officer
injuries column.

202

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Citizen Complaint Data
In Chapter 7 we examined citizen complaint data in order to assess the different use of
force policy approaches utilized by the eight agencies. In doing so, we analyzed external
complaints, with a focus on two complaint types that relate most to the use of force (i.e.,
discourtesy and use of force allegations). Our analyses, across our sites, focused on explaining
complaint frequencies, dispositions, discipline, comparisons to workload measures (i.e., calls for
service, reported crimes, arrests, and force), totals across officers, and multivariate comparisons.
What follows is a summary of the key findings from each of these sections.
Frequency of Complaints
We begin by comparing cities’ frequency of complaints for the use of force and
discourtesy. What follows is how cities ranked in terms of the percentage of citizen complaints
for use of force and discourtesy (compared to all external complaints). Summary rankings are
based on more detailed tables presented prior in our citizen complaint Chapter 7.
Use of Force Complaints
Across our two-year study period of citizen complaints that were reported and
investigated across our eight cities, St. Petersburg had the fewest for the use of force.
Conversely, we find the highest percentage of citizen complaints for the use of force (of all
external complaints) in Fort Wayne. For these rankings, and the remaining in this section, lower
numbers reflect more positive outcomes (i.e., less complaints), while the opposite is true for
higher rankings.

203

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

Rankings:
1-St. Petersburg, 2-Albuquerque, 3-Columbus, 4-Portland,
5-Colorado Springs, 6-Knoxville, 7-Charlotte-Mecklenburg, 8-Fort Wayne.

Discourtesy Complaints
In terms of the percentage of external complaints for discourtesy, Albuquerque had the
lowest ranking of the eight cities. Once again, Fort Wayne had the highest ranking based on
their percentage of external complaints that were for discourtesy, and thus the least positive
outcome.
<

Rankings:
1-Albuquerque, 2. St. Petersburg, 3-Portland, 4-Charlotte-Mecklenburg,
5-Colorado Springs, 6-Columbus, 7-Knoxville, 8-Fort Wayne.

Summary
In an attempt to summarize use of force and discourtesy amounts, as indicated by the
percentage of all external complaints, Table 9-8 adds the previous two rankings in this section
together. This allowed us to assess how cities fared cumulatively, compared to one another, in
terms of one of our primary policy outcomes - frequency of complaints. Not surprisingly, Fort
Wayne ranked the poorest of all cities. On the other end of the spectrum was Albuquerque and
St. Petersburg, both of whom fared the best (i.e., lowest) with respect to citizen use of force and
discourtesy complaints.
<

Rankings:
1-Albuquerque, 1-St. Petersburg, 3-Portland, 4-Columbus,
5-Colorado Springs, 6-Charlotte-Mecklenburg, 7-Knoxville, 8-Fort Wayne.

204

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-8. Total Use of Force and Discourtesy Citizen Complaints Comparison Rank
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Complaint Type
Use of Force
Discourtesy

5
5

4
3

2
1

8
8

1
2

6
7

7
4

3
6

Total

10

7

3

16

3

13

11

9

Rank

5

3

1

8

1

7

6

4

*Lower ranks indicate less complaints, higher ranks indicated more complaints

205

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Disposition of Complaints
After detailing the frequency of complaints, we then turned to examinations of
dispositions. Across the cities, dispositions were categorized as “sustained,” “not sustained,”
“exonerated,” or “unfounded.” Of particular concern was comparing cities in terms of the
percentage of sustained allegations, which we noted was the exception rather than the norm. The
key findings are as follows:
<

In terms of all citizen complaints combined, we found that in no city did the
sustained amount reach 30 percent. That said, the most sustained complaints (i.e.,
the highest percentage) were found in St. Petersburg, while the fewest were found
in Columbus.

<

For use of force allegations, the percentage of sustained complaints was extremely
low, ranging from zero (in Fort Wayne and St. Petersburg) to roughly five percent
(in Albuquerque).

<

With respect to discourtesy allegations, we found that sustained dispositions were
much more likely than for the use of force (over five times the amount at the high
end). Columbus had the fewest percentage of sustained discourtesy complaints,
while Knoxville had the highest.

Disciplinary Actions of Complaints
Next, we examined disciplinary actions of citizen complaint allegations. In comparing
cities, we focused on formal discipline (i.e., written reprimands, suspensions, and terminations)
over less formal (i.e., supervisory action, command counseling, and “other”) or no sanctions.
Across all cities, we noted that formal discipline of citizen complaints was very rare. The key
findings are as follows:
<

In terms of all citizen complaints, it was not too surprising to see formal
discipline was rather infrequent (ranging from roughly two percent in Columbus
to 16 percent in St. Petersburg), given the relatively low levels of sustained
complaints.

206

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

The percentage of use of force allegations that resulted in some form of formal
discipline was extremely low. Across cities, Colorado Springs, Fort Wayne, and
St. Petersburg did not have a single force complaint that resulted in formal
discipline. On the high end, although just over one out of every thirty allegations,
was Charlotte-Mecklenburg.

<

Like the pattern noted for sustained complaints, the percentage of allegations for
discourtesy resulted in higher amounts (in nearly every city) of formal discipline
when compared to the use of force. As was the case for all allegations, Columbus
had the fewest percentage of their discourtesy complaints result in some form of
formal action, while Charlotte-Mecklenburg (as they did for use of force
allegations) had the most.

Complaints in Relation to Calls for Service, Reported Crimes, Arrests, and Force
Besides explaining frequencies (and percentages) of the complaint process from overall
totals to formal discipline, across cities, we also placed complaints within the context of a variety
of agency workload measures (as we did previously in this chapter for the use of force). That is,
we examined citizen complaints for the use of force and discourtesy in relation to calls for
service, reported crime, arrest, and use of force.
Use of Force Complaints
In assessing use of force citizen complaints, in relation to our workload measures (see
Table 9-9), the most favorable city was St. Petersburg, who ranked first across average
complaints per 1,000 calls for service, per 1,000 Part I Index crimes, per 1,000 arrests, and per
100 use of force reports. On the other end of the spectrum, and thus ranking last, was Colorado
Springs with the highest cumulative score.
<

Rankings:
1-St. Petersburg, 2-Knoxville, 3-Fort Wayne, 4-Charlotte-Mecklenburg,
5-Portland, 5-Albuquerque, 7-Columbus, 8-Colorado Springs.

207

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Discourtesy Complaints
Comparisons of discourtesy citizen complaints, per workload measures (see Table 9-10),
revealed once again that St. Petersburg fared the best. On the other hand, Columbus
cumulatively fared the poorest of all cities.
<

Rankings:
1-St. Petersburg, 2-Charlotte-Mecklenburg, 3-Knoxville, 4-Albuquerque,
5-Portland, 6-Fort Wayne, 7-Colorado Springs, 8-Columbus.

Summary
In summarizing external complaints for the use of force and discourtesy, per calls for
service, reported crime, arrests, and use of force, in both cases, St. Petersburg ranked the best of
all cities. The two cities that ranked the least positive were Colorado Springs and Columbus,
both of which were more likely to receive external use of force and discourtesy complaints,
based on the citizen-based workload measures.
Use of Force and Discourtesy Officer Totals
Our next analysis extended the examination of citizen complaints for force and
discourtesy to the officer level. As such, we examined the number of officers that comprised the
combined total number of citizen force and discourtesy complaints per city. We also calculated
the average number of complaints per officer. For both sections, we restricted our analysis to
instances where an individual officer was identified.
Complaints per Sworn Personnel
In deducing complaints per sworn personnel, we divided the number of individual
officers that comprised the total number of external force and discourtesy allegations (over the
two-year period) by the total number of sworn personnel.
208

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table 9-9. External Use of Force Complaint Rank to Calls for Service, Reported Crime, Arrests and Force Reports
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Calls
Crimes
Arrests
Force Reports

7
8
6
8

8
6
5
4

5
5
7
6

3
4
3
2

1
1
1
1

2
3
2
3

4
2
4
7

6
7
8
5

Total

29

23

23

12

4

10

17

26

Rank

8

5

5

3

1

2

4

7

*Lower ranks indicate less external force complaints, higher ranks indicated more force complaints

Table 9-10. External Discourtesy Complaint Rank to Calls for Service, Reported Crime, Arrests and Force Reports
Col. Sp.

Port.

Albq.

Ft. Wy.

St. Pete

Knox.

Char.

Colum.

Calls
Crimes
Arrests
Force Reports

6
7
7
8

8
6
6
5

4
3
5
3

5
5
4
2

2
2
1
1

2
4
3
4

1
1
2
6

7
8
8
7

Total

28

25

15

26

6

13

10

30

Rank

7

5

4

6

1

3

2

8

*Lower ranks indicate less discourtesy complaints, higher ranks indicated more discourtesy complaints

209

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

This revealed rather substantial variation across cities. The fewest overall percentage of sworn
personnel with a citizen complaint for force or discourtesy was found in St. Petersburg, while the
most (i.e., least favorable) was found in Colorado Springs.
<

Rankings:
1-St. Petersburg, 2-Charlotte-Mecklenburg, 3-Albuquerque, 4-Fort Wayne,
5-Knoxville, 6-Columbus, 7-Portland, 8-Colorado Springs.

Officer Complaint Averages
We extended the officer analysis to compare the average number of citizen complaints
for force and discourtesy, where an officer was named, across the officers that had at least one
complaint. The city that fared the best (i.e., fewest average complaint totals) was Fort Wayne,
which is interesting given that they had the highest percentage of their external complaints for
force and discourtesy. The highest average complaint totals for force and discourtesy was found
in Columbus, and thus faring the poorest of all cities in this respect.
<

Rankings:
1-Fort Wayne, 2-St. Petersburg, 3-Knoxville, 3-Charlotte-Mecklenburg,
5-Albuquerque, 6-Colorado Springs, 7-Portland, 8-Columbus.

Multivariate Analysis
Our final examination of external complaints for the use of force and discourtesy, utilized
a multivariate approach to explain both types of complaints, while controlling for citizen factors
(i.e., sex, race, and age). In doing so, our primary aim was to isolate the independent effects of
each city. Our analytical model, which combined citizen complaints for force and discourtesy
(as we did in the preceding section) used Fort Wayne as the reference category, given their
aggregate findings regarding citizen complaints (see Table 9-8). Our analysis revealed:

210

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

<

While the likelihood of receiving a complaint was statistically lower than Fort
Wayne in all cities, Albuquerque and St. Petersburg represented the cities that
contrasted Fort Wayne the most. In terms of our citizen controls, none were
statistically distinguishable in our model.

Summary
Finally, in an effort to summarize several analyses, we combine rankings from three
separate examinations of citizen complaints to reveal a cumulative ordering across cities. More
specifically, we add the following city rankings: total citizen force and discourtesy (from Table
9-8) , the two workload rankings (combining results from Tables 9-9 and 9-10), and the external
force and discourtesy complaints per sworn personnel (from page 209). Our combined rankings
reveal that St. Petersburg fared the most positive (i.e., combined score of 3), with respect to
citizen complaints for force and discourtesy, while Colorado Springs cumulatively fared the
poorest (i.e., combined score of 20).
Total Citizen Force/Discourtesy
Ranking
1-Albuquerque
1-St. Petersburg
3-Portland
4-Columbus
5-Colorado Springs
6-Charlotte-Mecklenburg
7-Knoxville
8-Fort Wayne

Combined Workload Measures
Ranking
1-St. Petersburg
2-Knoxville
3-Charlotte-Mecklenburg
4-Albuquerque
4-Fort Wayne
6-Portland
7-Colorado Springs
7-Columbus

Citizen Force/Discourtesy per Sworn
Ranking
1-St. Petersburg
2-Charlotte-Mecklenburg
3-Albuquerque
4-Fort Wayne
5-Knoxville
6-Columbus
7-Portland
8-Colorado Springs

Combined Citizen Force/Discourtesy
Ranking
1-St. Petersburg
2-Albuquerque
3-Charlotte-Mecklenburg
4-Knoxville
5-Portland
5-Fort Wayne
7-Columbus
8-Colorado Springs
211

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Civil Litigation Data
In Chapter 8 we examined civil litigation data to assess how these data may reflect on the
different policy approaches used by our eight agencies. This enterprise proved to be challenging
with respect to insufficient departmental data collection protocols and/or a city’s reluctance to
release such information to researchers.64 In fact, only four cities provided useable data (i.e.,
Colorado Springs, Portland, Fort Wayne, and St. Petersburg). In two of the cities (i.e., Knoxville
and Charlotte-Mecklenburg), officials attempted to provide us the requested data, but we could
not find a format that we sought. In the remaining two cities (i.e., Albuquerque and Columbus),
the city attorney declined to provide these data despite repeated requests over the course of more
than one year. We leave it to the reader to draw conclusions as to why this may be the case and
the implications the failure to disclose such data may have on the findings relating to where these
cities may have ranked.
Number of Force Related Lawsuits Filed
Colorado Springs and St. Petersburg had the least number of force related lawsuits filed,
during our two year study period, while Fort Wayne had the most. The city rankings are as
follows (total number of force related lawsuits filed in parentheses) :
<

Rankings:
1-Colorado Springs (N=3), 1-St. Petersburg (N=3),
3-Portland (N=48), 4-Fort Wayne (N=51).

64

See Archbold and Maguire (2002) for an excellent discussion concerning the challenges of civil litigation data
collection.

212

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Number of Force Related Lawsuits Settled or Lost
Colorado Springs had the least number of force related lawsuits settled or lost during our
two year study period, while Portland and Fort Wayne had the most. The city rankings are as
follows (total number of force related lawsuits settled or lost in parentheses) :
<

Rankings:
1-Colorado Springs (N=0), 2-St. Petersburg (N=2),
3-Fort Wayne (N=20), 3-Portland (N=20).

Monetary Payout for Force Related Lawsuits
Colorado Springs had the least amount of monetary payout for force related lawsuits,
during our two year study period, while Portland had the most. The city rankings are as follows
(monetary payout in parentheses) :
<

Rankings:
1-Colorado Springs ($0), 2-Fort Wayne ($113,899),
3-St. Petersburg ($145,899), 4-Portland ($967,656).

Summary
Finally, if we compare how each of the four agencies ranked cumulatively across the
three civil litigation indicators listed above (force lawsuits filed, force lawsuits settled/lost, and
force lawsuits monetary payout), we found that Colorado Springs fared the best.
Combined Civil Litigation Ranking65
1-Colorado Springs
2-St. Petersburg
3-Fort Wayne
4-Portland

65

For this ranking order we simply summed the ranking for force lawsuits filed, force lawsuits settled/lost, and force
lawsuits monetary payout.

213

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

CHAPTER 10
Discussion
When we undertook this project, we sought to answer a couple of vast empirical
questions. At the simplest level, we wanted to decipher what kinds of use of force policies were
operating across the United States, and which of these policies provide more beneficial
outcomes. With this is mind, we gear this final chapter toward both the practitioner and research
community.

Police Administrators
Results from our national mail survey showed the extent to which force continua are
incorporated into written use of force policies. Two predominant findings emerged from this
research. First, a large majority of police agencies still use a force continuum approach, mostly a
linear continuum. Second, there is no commonly used means of tactical placement on the
continuum (i.e., where agencies place various hands on and weapons force).
Police administrators of agencies that currently use a force continuum, or those thinking
about adopting a force continuum approach, should take solace in the fact that most of their
colleagues across the country view the use of a force continuum as a guide to better decision
making. Articles appearing mainly in trade journal outlets (see Aveni, 2003, Peters and Brave,
2006, Petrowski, 2002, and Williams, 2002) attempting to cause alarm concerning the potential
negative effects of using a continuum have not led many agencies to abandon the use of such.
Unfortunately, where agencies place the different force options on the continuum is
tremendously varied, especially in relation to chemical sprays and electronic devices.

214

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

In answering the second overarching empirical question (i.e., which use of force policies
provide more beneficial outcomes?), we relied on data gleaned from in-depth study of eight
police agencies. Deciphering which of the differing policies was better (or best) was multidimensional, as outlined in the details of Chapters 5 through 9. In the end, we found that some
of the policies, across our departments, were better than others for certain outcomes, while for
other outcomes they may not have fared as well.
Going back to the seminal work of James Q. Wilson (1968), police chiefs consider
various audiences when running their organizations. Some chiefs prioritize officer concerns, and
thus would be interested in our outcome findings that relate more toward things of interest to
officers such as their perceptions of policy assistance, clarity, and guidance, as well as injuries
they may receive when engaged in forceful encounters with citizens. This approach, which may
endear chiefs to officers who view him/her as a “cop’s cop,” is more in line with what McGregor
(1960) views as Theory Y management. On the other hand, some chiefs might view citizen
needs as most salient in considering the best policy and practices, and thus would regard
outcomes such as frequency of force against citizens, citizen injuries, and force/discourtesy
citizen complaints as the primary outcomes. Finally, some chiefs may prioritize the operation
and functioning of the department as a whole, placing officer and citizen concerns secondary.
Such an approach would be more in line with McGregor’s view of Theory X management, and
might value such organizational outcomes as complaints across sworn personnel, use of force
policy compliance, and lawsuits. What follows is a discussion of each of these concerns that
chiefs may consider,66 as they relate to our findings.
66

In no way is this to imply that priorities are mutually exclusive, as chiefs undoubtedly consider all of these arenas.
Instead, we suggest that chiefs may prioritize, or accentuate one over the other, as a function of their style.

215

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Officer Concerns67
A police executive management style that places primary importance on officers’ views
and beliefs may be particularly interested in our findings from the patrol officer survey. For
instance, we found that officers in some agencies believe their use of force policy impacts the
likelihood of citizen/officer injuries, complaints, and lawsuits. Officers in Knoxville,
Albuquerque, and Colorado Springs were more likely than officers in other agencies to believe
that their policy approach helps decrease citizen/officer injuries. Conversely, modal survey
responses indicated that St. Petersburg and Charlotte-Mecklenburg officers believe their use of
force policy actually increases the likelihood of officer injuries. Interestingly, officers in both of
these agencies also felt that their policy was overly restrictive when compared to officer views in
our other agencies. Thus, St. Petersburg and Charlotte-Mecklenburg officers may feel that an
unduly restrictive policy leads to an enhanced risk of injuries.
Police administrators may also draw on our analysis involving the actual reported number
of officer injuries in relation to the total number of force reports filed in each agency over the
course of our two year study period. In this case, we found that Columbus and Portland fared
best (i.e, had the least officer injuries), while Knoxville and Charlotte-Mecklenburg fared worse
(i.e., had the most officer injuries). If these actual officer injury rates are considered in tandem
with the survey findings on officer views of injuries, officials in Charlotte-Mecklenburg may
wish to re-evaluate their policy approach. In short, patrol officers in this city not only believe
their policy subjects them to more injuries, the statistical analysis on the amount of injuries bears
this perception out in reality. On the other hand, while St. Petersburg officers felt their policy
67

For the officer, citizen, and departmental concerns that follow, we provide illustrative (but not exhaustive)
examples of administrator considerations throughout. For a more detailed discussion of these outcomes, we refer the
reader back to the individual chapters (i.e., five through eight), as well as the project summary findings in Chapter 9.

216

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

leads to more injuries, our analyses does find this to be the case. In fact, St. Peterburg had the
third lowest rate of injuries. In this case, officials may simply want to meet with officers to call
attention to this disconnect and reassure them that their rate of injuries is actually relatively low.
Citizen Concerns
Police managers primarily concerned with how citizens perceive police use of force
behavior may be less interested in patrol officer perceptions and injuries, and more interested in
some of our other indicators. For example, both Colorado Springs and Knoxville fared well
when considering how often officers used force in relation to a combination of various workload
measures (i.e., calls for service, reported crimes, and arrests). In other words, officers in these
cities, in general, relied on less force to accomplish their means.68 However, Portland and Fort
Wayne officers were at the opposite end of the spectrum. Officers in these cities tended to rely
on more coercive means during their encounters with citizens. As a result, police executives
more sensitive to potential citizen concerns, with respect to the frequency with which officers
use force, may wish to consider policies that mirror those of Colorado Springs and Knoxville,
and shy away from the approach taken in Portland and Fort Wayne.
Another indicator a pro-citizen police administrator may consider is citizen complaints.
When we compared citizen complaints for use of force and discourtesy, in relation to
documented use of force, St. Petersburg and Fort Wayne fared the best. As such, administrators
who might prioritize citizen complaints, per use of force encounters with the public, might

68

This is to not say that certain forms of force that some may consider to be fairly coercive in nature were always
used infrequently. For example, Colorado Springs reported using a CED in half of all their police-suspect force
encounters (49.5%). The agency with the next highest frequency of use was Portland at 23.4 percent, or less than
half what Colorado Springs reported. Nonetheless, when workload measures, and the extent to which Colorado
Springs officers faced resistant suspects, are factored into the equation, the frequency of force usage is actually
rather low compared to officers in the other cities.

217

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

consider adopting policies found in these two cities, while they might be more critical of
approaches endorsed by Colorado Springs and Charlotte-Mecklenburg - both of whom fared
poorly on this outcome. Of interest, is the perception-behavior accuracy and disconnect noted
among two of the aforementioned agencies. That is, St. Petersburg and Colorado Springs were
the only two agencies where the modal response by officers was that their use of force policy
increases the potential for citizen complaints. The behavioral data on complaints revealed that
Colorado Springs officers were accurately perceiving their policy’s influence, while St.
Petersburg was not.

Overall Departmental Concerns
Just as some chiefs may prioritize officer or citizen concerns, in defining their
management style, others might view the functioning of the department as a whole as their
primary concern. Should this be the case, one outcome of concern may be the percentage of
citizen complaints for force and discourtesy in relation to the total number of sworn officers in a
department. St. Petersburg and Charlotte-Mecklenburg, with approximately 10 percent of their
personnel named in a complaint, ranked the most favorably, while Portland and Colorado
Springs, with over two and a half times the amount, fared the worst. As a model for police
chiefs, the policy approaches by St. Petersburg and Charlotte-Mecklenburg would appear to be
the more exemplary, than the rest, for this outcome.
Another outcome, for executives with a management style that focuses on the concerns of
the organization as a whole, is our analysis of lawsuits. While these data are limited to half of
our agencies, Colorado Springs fared the best across the board (i.e., number of force lawsuits

218

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

filed, settled/lost, and payouts). With respect to two indicators of primary importance for police
leaders (i.e., lawsuits settled/lost and payouts), Portland fared the worst. In connecting lawsuits
to use of force, it might be of interest to police chiefs that Colorado Springs and Portland were
on the opposite end of the spectrum in terms of force, per our workload measures. As such, the
organization that fared the best in terms of lawsuits generally relied on force less often to
accomplish their goals, while the department that fared the worst with respect to lawsuits,
generally used force more readily.

Police Researchers
Besides police administrators, another primary audience of this study is the research
community. Across the four years of this project, we experienced a number of challenges and
lessons. Below we highlight a few, which might assist those who evaluate this work and/or plan
on undertaking a similar research endeavor. We also explain some of the limitations of our
project, as well as what we would like to see in terms of future use of force research.

Study Challenges
Throughout the course of the past four years we faced a number of challenges. Here we
offer some reflections in reference to these challenges. Our primary intent is to offer the reader a
better sense surrounding the “nuts and bolts” of this type research endeavor; but we also hope
that other researchers may glean something from these challenges that may assist in their studies.
Our initial challenge occurred during the transitional stage between Phases I and II.
Because we wanted to select eight agencies for in-depth examination, based on the most

219

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

frequently used policy approaches nationally, we were unable to identify and secure these
agencies until Phase I of the project (i.e., the agency survey) was completed. This created a time
lag between the two phases. In addition, we also had stringent agency selection criteria that
slowed the process. While a convenience sample of agencies would have been much easier (i.e.,
selecting local agencies or departments we worked with in the past), we feel the approach taken
offers more empirically valid and reliable results.
Once the eight agencies were selected, we were faced with a host of additional
challenges. We highlight but a few of them here. First, since we were in the field for two years,
personnel turnover was sometimes an issue. In one of our cities, the police chief who originally
agreed to the study stepped down, and we had to seek approval with the new incoming chief to
ensure continued participation. Relatedly, our key agency contacts or liaisons assigned to us
were sometimes reassigned during the project, requiring us to begin anew with their
replacements.
Second, because of the numerous data elements involved in the project, we sometimes
had trouble identifying (or finding) the “right person” in reference to certain pieces of
information we sought. At times, this person (on paper) was not the one you would have thought
could help you obtain the data that you wanted. The key here is moving around, when
applicable, and going beyond departmental listings of personnel to get in touch with the person
who has access to (and can get) the data needed.
Third, because we wanted to survey the population of police officers assigned to patrol
assignments, rather than selecting a sample of officers, this part of the project required a great
deal of time and effort. This meant that project staff was generally on-site for a week to ten days

220

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

so as to visit every patrol shift, across each geographic location, at least twice. In some cases,
we attended five different roll call sessions a day. While such an effort was certainly laborious,
we feel the strengths gained (i.e., valid and reliable data, extremely high response rates) far
outmatch some of the other survey methods used by others (e.g., dropping surveys off to have
agency personnel distribute, internet surveys, etc.).
Finally, it was imperative that we develop “common measures” of officer force and
subject resistance (as a foundation of this grant) because the agencies did not always use the
same definitions or terminology. This often required us to have to go back and re-code these
measures based on the descriptions offered in the force reports. In some cases, we had to request
hard copies even when much of the data were already captured electronically. This slowed the
project, but once again should reflect the confidence we have in our findings.
Study Limitations
While the present study has many strengths, as demonstrated throughout this report, there
are also a number of limitations. For example, two of our primary data sources were surveys
(i.e., agency and patrol officer). Given that use of force inquiries at both levels of analysis are
rare, we were without established measures to apply to our study. As such, in several instances,
we constructed our own questions, and even though they were pre-tested on a number of police
personnel from various agencies, our confidence in these measures will be enhanced once they
are replicated across other research sites.
In addition to the limitations faced with our two surveys, the use of official records
(which provided us with the bulk of our project data) also came with a number of drawbacks.
First, official records are documented by an “interested party” with a stake in the outcome. For

221

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

instance, officers filling out use of force reports may do so in a manner that illustrates favorable
behavior on their part or ex post facto justifications for behavior (e.g., CYA), while the same
holds for the citizen who recalls events that occur between themselves and the police. So,
whether it be the officer documenting the force used against a citizen or the citizen filling out a
complaint against an officer, the varying perspectives of official documentation per each actor
suggests that the truth of how police-citizen encounters occur probably lies somewhere between
the officer’s account of the event and the citizen’s version.
At the same time, as social scientists, we have to consider that police who are
documenting behavioral documentation are not as concerned as researchers might be in the
completeness of the data. For example, in cases where officers were afforded, per the official
use of force report form, a series of boxes to check for each type of force utilized (e.g., firm grip,
escort, pressure point, take down, punch, kick, etc) it is conceivable that in instances where
higher levels of physical force (e.g., kick) were displayed along with a lower level of force (e.g.,
firm grip) that only the higher level of force was checked off and documented.
In a similar vein, instances in official documentation (e.g., use of force, officer injury,
citizen injury, etc.) where officers were provided a response category of “other,” which could
include a host of behaviors, we found that sometimes there was further explanation in an open
ended fashion and at other times the box was just checked “other.” The latter approach did not
provide us the opportunity to recode the “other” into a discrete classification of behavior.
Another limitation involved determining the temporal ordering of force and resistance within
individual encounters, an issue discussed in previous works (Terrill, 2001, Terrill and
Mastrofski, 2002). The level of detail provided by officers in their narrative descriptions

222

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

sometimes left us uncertain as to the specific sequencing of resistance and force behaviors.
Thus, as indicated in Chapter 6, the reader is cautioned as per our description of officers using
force within the context of policy.
Similarly, citizens filing complaints do so through their own perspective, and like
officers, without the cental concern of establishing quality official data. For example, there were
instances where citizens filed a complaint against the police in general, without a specific officer
named, which affected our ability to calculate precise officer-level totals. This forced us to
exclude these complaints (which in one of our sites was fairly substantial) from our analyses in
Chapter 7.
Additional data limitations were a function of various organizational layers beyond patrol
officer and citizen documentation. In some cases the availability of official records was absent
or so lacking that we could not conduct a comprehensive data analyses. For instance, none of
the cities captured suspect wealth or demeanor, both of which have been shown to be fairly
consistent predictors of police use of force behavior. One of our research sites also failed to
capture citizen race on their use of force forms, another correlate of police use of force. Further,
as illustrated in Chapter 8, the civil litigation data were only provided and usable in four of the
eight cities.
Finally, while we believe that the eight mid-to-large sized agencies in the present study
offer a nice degree of generalizability, our findings may not translate to all types of police
departments. That is, the results of this project may not be generalizable to smaller rural
agencies or very large urban deparments where the dynamics of use of force policies and
outcomes may be operating in a different manner.

223

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Future Research
We conclude with a few recommendations for future research. First, researchers should
continue to examine which policies appear to work best. As documented throughout this report,
what works best really depends on what perspective one comes from and the indicator one
considers. Thus, future research should determine whether the findings reported here are
replicated in other settings. Moreover, researchers may develop their own approaches for how to
rank agencies based on a multitude of different indicators.
Second, we found that most agencies use a force continuum within their written policy
and that they overwhelming select a linear design. This begs for further inquiry. For instance,
one may query police executives as to “why” they so often incorporate such an approach and the
perceived benefits and drawbacks. Alternatively, one may examine why the placement of
chemical sprays and CEDs seem to pose the greatest difficulty in terms of tactical placement.
Third, researchers need to be ever conscience as to the difficulty of measuring and coding
use of force behavior via the use of official records, especially use of force reports. We created
common measures of both officer force and citizen resistance and would strongly advise future
researchers to consider some form of a similar approach. Without the use of common measures,
the ability to compare across agencies becomes a seriously challenging, if not impossible, task.
Finally, with regard to collecting data, especially from police departments, researchers
should not be afraid to “get dirty.” Data that are easy to collect is always a bonus, but in some
circumstances may not provide the most reliable and valid information. Thus, a good researcher
is often faced with a balancing act between getting the highest quality data and not burning out.
From the agency mail survey (i.e., individually calling departments), to picking comparable (but

224

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

different) agencies to study, to convincing departments that certain data did not need to be
redacted, to surveying a population of line officers about their policy, to photo copying and
coding (by hand) from hard copy files, to sitting with data analysts and trying to figure out how
to get “data fields” into individual data, to coming up with like measures for comparisons, there
were certainly more convenient ways for us to go at various points over the past four years.
Much of this research had to be done in the field and during site visits that required sustained
and substantive periods of work! Too often, in our experiences, researchers get side tracked with
finding out how to easily get the data (e.g., often incomplete electronic data), having dinner with
the police chief, or engaging in politics in an attempt to secure the next consulting contract.
Simply put, we implore our research colleagues to go the extra mile to obtain the best data
possible.

225

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

REFERENCES
Adang O. and J. Mensink (2004). “Pepper Spray: An Unreasonable Response to Suspect Verbal
Resistance.” Policing: An International Journal of Police Strategies and Management 27
(2): 206-219.
Alpert, G.P. and R.G. Dunham (1997). The Force Factor: Measuring Police Use of Force
Relative to Suspect Resistance. Washington, DC: Police Executive Research Forum.
Alpert, G.P. and R.G. Dunham (2004). Understanding Police Use of Force: Officers, Suspects,
and Reciprocity. New York, NY: Cambridge University Press.
Amnesty International (2004). “Excessive Use of TASERs in USA.” The Wire (November).
Archbold, C.A. and E.R. Maguire, E.R. (2002). “Studying Civil Suits Against the Police: A
Serendipitous Finding of Sample Selection Bias.” Police Quarterly 5 (2): 222-249.
Aveni, T.J. (2003). “Force Continuum Conundrum.” Law and Order 51 (12): 74-77.
Bertomen, L. (2003). “Less-Lethal Toolbox: Law Enforcement Officers Test the Products on the
Use-of-Force Continuum.” Law Enforcement Technology 30 (10): 64-69.
Cheh, M. M. (1995). “Are Lawsuits an Answer to Police Brutality?” In W. Geller and H. Toch
(Eds.), And Justice for All: Understanding and Controlling Police Abuse of Force (pp.
233-259). Washington, D.C.: Police Executive Research Forum.
Colorado Spring Police Department (CSPD), General Order 705: Use of Force Continuum.
Colorado Spring Police Department (CSPD), Standard Operating Procedure P1-171: Conducted
Energy Devices.
Connor, G. (1991). “Use of Force Continuum: Phase II.” Law and Order 39 (3): 30-32.

226

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Crawford C. and R. Burns (1998). “Predictors of the Police Use of Force: The Application of a
Continuum Perspective in Phoenix.” Police Quarterly 1(4): 41-64.
Dillman, D. (1978). Mail and Telephone Surveys: The Total Design Method. New York: Wiley.
Federal Bureau of Investigation, United States Department of Justice (October 2008).
Law Enforcement Officers Killed and Assaulted, 2007. Retrieved (May 22, 2011), from
http://www.fbi.gov/ucr/killed/2007/.
Fyfe, J. J. (1986). Police Personnel Practices—1986, Baseline Data Reports (Vol. 18, No. 6).
Washington, DC: International City Management Association.
Garner, J.H., Schade, T., Hepburn, J., and J. Buchanan (1995). “Measuring the Continuum of
Force Used by and Against the Police.” Criminal Justice Review 20: 146-168.
Graham v. Connor, 490 U.S. 386 (1989).
Hickman, M. and B. Reaves (2003). Law Enforcement Management and Administrative
Statistics: Local Police Departments 2000. Bureau of Justice Statistics.
Hoffman, R., Lawrence, C., and G. Brown (2004). “Canada’s National Use-of-Force Framework
for Police Officers.” The Police Chief 71 (10).
Kaminski, R.K., Edwards, S.M., and J.W. Johnson (1999). “Assessing the Incapacitative Effects
of Pepper Spray During Resistive Encounters with the Police.” Policing: An
International Journal of Police Strategies and Management 22 (1): 7-29.
Klinger, D.A. (1995). “The Micro-Structure of Nonlethal Force: Baseline Data from an
Observational Study.” Criminal Justice Review 20: 169-186.

227

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Lersch, K.M. and T. Mieczkowski (2000). “An Examination of the Convergence and Divergence
of Internal and External Allegations of Misconduct Filed Against Police Officers.”
Policing: An International Journal of Police Strategy and Management 23: 54-68.
McEwen, T. (1997). “Policies on Less-than-Lethal Force in Law Enforcement Agencies.”
Policing: An International Journal of Police Strategy and Management 20 (1): 39-59.
McGregor, D. (1960). The Human Side of Enterprise. New York: McGraw-Hill Publishing.
Paoline, E.A. III and W. Terrill (2004). “Women Police Officers and the Use of Coercion.”
Women and Criminal Justice, 15 (3/4): 97-119.
Paoline, E.A. III and W. Terrill (2007). “Police Education, Experience, and the Use of Force.”
Criminal Justice and Behavior, 34 (2): 179-196.
Pate, A.M. and L.A. Fridell (1993). Police Use of Force: Official Reports, Citizen Complaints,
and Legal Consequences. Washington, D.C.: The Police Foundation.
Peters, J and M. Brave (2006). “Force Continuums: Are they Still Needed.” Police and Security
News 22 (1): 1-5.
Petrowski, T.D. (2002). “Use-of-Force Polices and Training: A Reasoned Approach.” The FBI
Law Enforcement Bulletin (October).
Rogers, D. (2001). “Use of Force: Agencies Need to Have a Continuum and Officers Need to Be
Able to Articulate It.” Law Enforcement Technology 28 (3): 82.
Smith, M.R. and G. P. Alpert (2000). “Pepper Spray: A Safe and Reasonable Response to
Suspect Verbal Resistance.” Policing: An International Journal of Police Strategies and
Management 23 (2): 233-245.

228

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Smith, M.R., Kaminski, R.J., Rojek, J., Alpert, G.P., and J. Mathis (2007). “The Impact of
Conducted Energy Devices and other Types of Force and Resistance on Officer and
Suspect Injuries.” Policing: An International Journal of Police Strategies and
Management 30: 423-446.
Sykes, R.E. and E.E. Brent (1983). Policing: A Social Behaviorist Perspective. New Brunswick,
NJ: Rutgers University Press.
Terrill, W. (2001). Police Coercion: Application of the Force Continuum. New York: LFB
Scholarly Publishing.
Terrill, W. (2003). “Police Use of Force and Suspect Resistance: The Micro-Process of the
Police-Suspect Encounter.” Police Quarterly 6 (1): 51-83.
Terrill, W. (2005). “Police Use of Force: A Transactional Approach.” Justice Quarterly 22 (1):
107-138.
Terrill, W. and S. D. Mastrofski (2002). “Situational and Officer Based Determinants of Police
Coercion.” Justice Quarterly 19: 215-248.
Terrill, W. and J. McCluskey (2002). “Citizen Complaints and Problem Officers: Examining
Officer Behavior.” Journal of Criminal Justice 30: 143-155.
Terrill, W. and E.A. Paoline III (2010). “Non-Lethal Force by Police: The Various Lenses by
Which Appropriateness is Assessed.” In J.B. Kuhns and J. Knutsson (Eds.), Policing
Around the World: Police Use of Force, Firearms, and Non-Lethal Weapons (pp. 6-13).
Westport, CT: Praeger-Greenwood-PSI.
Terrill, W., Paoline, E.A. III, and P.K. Manning (2003). “Police Culture and Coercion.”
Criminology, 41 (4): 1003-1034.

229

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Terrill, W., Alpert, G.P., Dunham, R.D., and M.R. Smith (2003). “A Management Tool for
Evaluating Police Use of Force: An Application of the Force Factor.” Police Quarterly 6
(2): 150-171.
United States President's Commission on Law Enforcement and Administration of Justice
(1967). Task Force on the Police. Washington, D.C.: Government Printing Office.
Weisburd, D., Greenspan, R., Hamilton, E.E., Williams, H., and K.A. Bryant (2000). “Police
Attitudes Toward Abuse of Authority: Findings From a National Study.” National
Institute of Justice Research in Brief (May).
Williams, G.T. (2002). “Force Continuums: A Liability to Law Enforcement?” The FBI
Law Enforcement Bulletin (June).
Wilson, J. Q. (1968). Varieties of Police Behavior: The Management of Law and Order in Eight
Communities. Cambridge, MA: Harvard University Press.

230

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Appendix A

NATIONAL SURVEY OF POLICE AGENCIES:
Examining Force Policy Types

Conducted by:

Dr. William Terrill

MICHIGAN STATE UNIVERSITY
School of Criminal Justice
and

Dr. Eugene A. Paoline III

UNIVERSITY OF CENTRAL FLORIDA
Department of Criminal Justice

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

General Project Information
In the Fall of 2005, the National Institute of Justice (NIJ) funded researchers from Michigan
State University and the University of Central Florida to conduct a study examining the varying
types of less-than-lethal force policies currently being used throughout the country. The study
seeks to identify the extent of variation in force policies. Your participation in this study will
provide valuable insight regarding the manner in which most police officers are instructed on the
use of force and lead to informed policy understanding and development.
We seek your cooperation in gathering this information by completing the survey, which is
expected to take approximately 15 minutes. Your participation is voluntary and your answers
are strictly confidential. Your confidentiality will be protected to the maximum extent allowable
by law as guaranteed by federal statue (42 United States Code 3789g). You may choose not to
participate at all, refuse to participate in certain procedures or answer certain questions, or
discontinue participation at any time without penalty.
There are no known risks associated with participation in this study. Only the research team will
see your responses and the survey will be destroyed upon completion of the project at the end
of 2007. The identification number on the back of the survey is used solely to remove your
agency from the mailing list to avoid future mailings. The information collected will be compiled,
analyzed, and presented only in grouped form with no individual agency identified.
If you have any questions about this study please contact the principal investigators, Professor
William Terrill, Michigan State University, School of Criminal Justice, 532 Baker Hall, East
Lansing, MI 48824, 517-353-9752 (phone), 517-432-1787 (fax), terrillw@msu.edu (email) or
Professor Eugene A. Paoline III, University of Central Florida, P.O. Box 161600, Orlando, FL
32816, 407-823-4946 (phone), 407-823-5360 (fax), epaoline@mail.ucf.edu (email). If you have
questions or concerns regarding your rights as a study participant, or are dissatisfied at any time
with any aspect of this study, you may contact (anonymously, if you wish) Peter Vasilenko,
Ph.D., Chair of the University Committee on Research Involving Human Subjects, 202 Olds
Hall, East Lansing, MI 48824, 517-355-2180 (phone), 517-432-4503 (fax), ucrihs@msu.edu
(email) or Barbara Ward, Institutional Review Board (IRB), University of Central Florida, 12443
Research Parkway, Suite 302, Orlando, FL 32826, 407-823-2901 (phone), 407-823-3299 (fax),
bkward@mail.ucf.edu (email).
You indicate your voluntary agreement to participate in this research and have your answers
included in the data set by signing below and completing the survey. Thank you for your
assistance in this effort to enhance policy understanding and development.

PLEASE SIGN HEREÆ _________________________
Respondent Signature

2

________________
Date

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

SECTION I. In this section we would like to learn about your agency’s less-thanlethal force policy, including the types of force permitted or allocated and the
existence of specific written directives. We ask that you circle your response.
1.

Does your agency have a written policy on the use of less-than-lethal force?
[1] Yes
[2] No (please go to Section III on page 8)

2.

Please identify the varying forms of force that your agency permits or allocates
AND whether there are specific written directives in your use of force policy
detailing how and/or when to use such force. By specific written directives, we
mean that your policy directly discusses the application of individual force types.
Agency Permits
Force Type
or Allocates
Officer presence…………………………………….
Yes No

Covered in
Written Policy
Yes No

Verbal direction, other than commands/threats....
(for example - questioning)

Yes No

Yes No

Verbal commands/threats………………………….
(for example - orders, threats)

Yes No

Yes No

Soft-empty hand techniques……………………….
(for example - grabbing, shoving)

Yes No

Yes No

Pain compliance techniques……………………….
(for example - pressure point control)

Yes No

Yes No

Hard-empty hand techniques………………………
(for example - striking with fists)

Yes No

Yes No

Handcuffing…………………………………………..

Yes No

Yes No

Other forms of cuffing……………………………….
(for example - legs)

Yes No

Yes No

Flashlight……………………………………………..

Yes No

Yes No

Baton………………………………………………….
(for example - ASP)

Yes No

Yes No

Chemical-irritant sprays…………………………….
(for example - oleoresin capsicum)

Yes No

Yes No

3

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agency Permits
or Allocates

Force Type

K9………………………………………………………

Covered in
Written Policy

Yes No

Yes No

TASER………………………………………………… Yes No

Yes No

Electronic devices……………………………………
(other than TASER)

Yes No

Yes No

Projectile launchers………………………………….
(for example - beanbag)

Yes No

Yes No

Shotgun or automatic rifle…………………………..

Yes No

Yes No

Handgun……………………………….……………..

Yes No

Yes No

Other (please list all the other types of force that are permitted or allocated
and/or for which there is specific written direction concerning such force):
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________

SECTION II. In this section we would like to learn whether your agency’s lessthan-lethal force policy involves the use of a “force continuum,” and if so, about
the type of continuum approach used, how the varying forms of resistance/force
are ranked on the continuum from least to most severe, and how your continuum
links corresponding force options to varying levels of citizen resistance.
1.

Is a “force continuum” approach used in your policy on less-than-lethal force?
By force continuum, we mean a guideline (sometimes depicted graphically) that
officers can use to determine the type of force that may be used in generic
situations. Such guidelines are sometimes (but not always) linked with varying
forms of citizen resistance in an attempt to assist officers in matching the level of
force to the level of resistance/threat encountered. Some examples include linear
(e.g., ladder, stair, FLETC), wheel, and matrix/box designs, although there are
many variations of continuum designs besides these few examples.
[1] Yes, this agency uses some form of a force continuum.
[2] No, this agency does not use a force continuum (please go to Section III on
page 8).
4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

2.

Identify the “type” of force continuum your agency uses. Please understand that
there is no correct or ideal continuum design in existence. Rather, agencies
across the country simply prefer one design rather than another. To help guide
you, we offer basic examples of some of the continuum designs currently in
existence, but we would expect your agency’s continuum to vary to some extent
(for example, in terms of the number and location of resistance/force options).
With this in mind, please circle ONE of the following five continuum designs (A-E)
that is similar (but may not necessarily be an exact match) to your agency’s
continuum. If your agency’s continuum design is not similar to one of the five
examples, we ask that you circle F and assist us in understanding your policy.
[A] Linear design (e.g., step, ladder, stair) WITHOUT graphic representation
For example:
Citizen Resistance/Threat Level
Assaultive (deadly)
Assaultive (non-deadly)
Active
Passive
Compliant

Corresponding Officer Force
Deadly force
Striking maneuvers, electrical devises
Pain compliance, chemical sprays
Verbal commands, soft-empty hand
Presence, verbal direction

[B] Linear design (e.g., step, ladder, FLETC model) WITH graphic representation
For example:

5

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

[C] Partial Wheel/Circular design
For example:

[D] Full Wheel/Circular design
For example:

6

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

[E] Matrix/Box design
For example:

[F] Other - my agency uses another type of continuum design (please describe
here below and offer a graphic example if applicable):

7

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

3.

Please identify the progression of citizen resistance and police use of force
outlined in your continuum policy, from the lowest level to the highest level, and
indicate the types/levels of force recommended to officers for each type of
resistance/threat they encounter (i.e., the range of force options available to
officers for each type of citizen resistance encountered). For example:
Citizen Resistance/Threat Level

Corresponding Officer Force

Assaultive – deadly

Deadly force

Assaultive - non-deadly
(e.g., punching, kicking)

Striking maneuvers, electrical devices
(e.g., hand strike, baton, TASER)

Active
(e.g., fleeing, pulling away)

Pain compliance, chemical sprays
(e.g., wristlock, OC spray)

Passive
(e.g., ignoring, refusing order)

Verbal threats, soft-empty hand
(e.g., firm grip)

Compliant

Presence, verbal direction

Please list below how your agency ranks and links resistance/force:
Citizen Resistance/Threat Level

Corresponding Officer Force

_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________
_______________________________________________________________

8

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

4.

Does your continuum policy “specifically” take into consideration any factors
other than citizen resistance (for example - suspect height/weight, perceived
mental state, drug/alcohol use, seriousness of offense, presence of weapon,
etc.)? If so, please state what factors are considered:
____________________________________________________
____________________________________________________
____________________________________________________
____________________________________________________
____________________________________________________
____________________________________________________

SECTION III. In this section we would like to learn about your policies on use of
force reporting.

1.

Does your agency currently require officers or supervisors to file a “use of force
report” (or other similar report such as a “control persons” or “supervisor control
persons” report) when officers use force or citizens complain of injury?
[1] Yes
[2] No (please go to Section IV on page 10)

2.

Who is primarily responsible for completing the report?
[1] Officer
[2] Supervisor
[3] Other personnel (please identify________________________)

3.

At what level of force is an officer or supervisor required to file a force report?
For example, some agencies require officers to file a force report for any type of
physical force (including simple restraint, handcuffing, and come-alongs). Other
agencies require a force report only when officers use a level of force that goes
above and beyond simple restraint. Yet, other agencies use different reporting
requirements. Please identify at what level of force your agency requires officers
file a force report:
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
9

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

4.

Is an officer or supervisor required to file a force report whenever a citizen
complains of injury, regardless of the level of force the officer used?
[1] Yes
[2] No

5.

As part of your reporting system, is a supervisor required to go to the scene?
[1] Yes
[2] No (please skip go to question 7)

6.

Is the supervisor required to interview concerned parties (for example - victim,
witnesses, officers)?
[1] Yes
[2] No

7.

Are force reports routinely reviewed by an immediate or shift supervisor (for
example - a shift Sergeant or Lieutenant conducting a review of the report prior to
the end of the shift)?
[1] Yes
[2] No

8.

Are force reports routinely reviewed beyond the immediate or shift supervisor (for
example - Internal Affairs or the Training Unit conducting a monthly review of
force reports)?
[1] Yes
[2] No

9.

Did your agency require officers or supervisors to fill out a “use of force report” in
2005 and 2004?
[1] Yes
[2] No

10.

Has your agency altered/changed it’s policy on use of force reporting (for
example, changing the criteria for when officers are required to fill out a form)
within the past two full calendar years (i.e., 2005 or 2004)?
[1] Yes (please list the approximate date the change occurred_____________)
[2] No

10

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

SECTION IV. In this section we would like to learn a little about your organization
and workload measures, which will help us to accurately report results with those
from similar types of agencies.

1.

Number of sworn officers (as of today): __________________

2.

Number of civilians (as of today):__________________

3.

Please indicate the number of “calls for service” your agency received during the
2004 calendar year:
Total number of calls for service: ____________

4.

Please indicate the number of “reported crimes” your agency received during the
2004 calendar year for each of the following categories:
Violent Crimes

# of crimes during the 2004 calendar year

Homicide
Rape
Robbery
Aggravated Assault

_____________
_____________
_____________
_____________

Property Crimes

# of crimes during the 2004 calendar year

Burglary
Larceny Theft
Motor Vehicle Theft
Arson

_____________
_____________
_____________
_____________

SECTION V. Please offer any other information on your use of force policy that
you believe may be helpful.
Comments on use of force policy:
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
________________________________________________________________
11

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

In conclusion, we ask that you PLEASE include 1) a copy of your less-than-lethal force
policy when returning this survey, and 2) a business card (or list contact information
below) of the person who filled out the survey. We seek this information for clarification
purposes only. As mandated by federal law (42 United States Code 3789g), we are
prohibited from releasing such information to anyone outside the research team. These
materials will be securely discarded upon the completion of the project.
Name_________________________________________________________________
Rank/Title______________________________________________________________
Agency________________________________________________________________
Address_______________________________________________________________
City____________________________ State_______ Zip________________________
Phone_____________________________________ Fax________________________
Email_________________________________________________________________
In your estimation, how receptive would your agency be toward being included in a
more in-depth analysis of less-than-lethal force policy and practice, as part of a select
pool of contemporary American police agencies?
[1] Very receptive
[2] Somewhat receptive
[3] Not sure
[4] Somewhat unreceptive
[5] Very unreceptive

Thank you for taking the time to fill out this survey. Your input will help shed light on
existing policy approaches. If you have not signed the consent form on page 1, we
ask that you please do so now in order for us to comply with federal research
regulations. As previously noted, the identification number below is used solely to
remove your agency from the mailing list so as to avoid future mailings. Your answers
are strictly confidential as ensured by federal law.
Please return survey to:
Dr. William Terrill,
Michigan State University,
School of Criminal Justice,
532 Baker Hall,
East Lansing, MI 48824

ID #:
12

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
Appendix
B of the U.S. Department of Justice.
and do not necessarily reflect the official position
or policies

Columbus
Action-Response Levels of Control:
Level 1: Empty hand control, pressure points, grounding techniques,
and joint manipulations
Level 2: Use of chemical spray
Level 3: Use of electronic device (electronic custody belt or taser)
Level 4: Hard empty hand control (strike/punch/kick)
Level 5: Use of impact weapon (baton/flashlight)
Level 6: Police K-9 bite
Level 7: Less lethal weapons (beanbag/multiple baton rounds/stinger cartridges)
Level 8: Deadly force

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Charlotte-Mecklenburg Police Department
Interactive Directives Guide

Levels of
Resistance
Levels of
Control

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Portland

LEVEL OF
CONTROL

METHOD OF
CONTROL

SUBJECT' S
BEHAVIOR

Mere presence

Officer ' s presence

COlnpliance

Verbal control

Verbal request
Questioning order

COlnpliance
Verbal n oncoHlpliance

Physical control

Control holds
Pressure point control

Passive resistance

Aerosol restraint
Taser: Touch Stun
Taser: Probes

Physical resistance
Indicates intent to
engage in physical
resistance.

Baton
Strikes & Kicks

Aggressive physical
resistall_ce.
Indicates intent to
.
.
engage 111 aggreSSIve
physical resistance.

Less lethal specialty
ITlunitiolls.

Aggressive physical
re sistance.
Indicates intent to
.
.
engage III aggreSSIve
physical resistance.
Anned or potentially
anned, capable of
causing serious physical
injury or death.

FirearlTlS

Deadly force

IInpact "\N'eapons.

Deadly force

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Albuquerque
REACTIVE CONTROL MODELM(RCMr
SUBJECT BEHAVIOR:

NON-COOPERATIVE

SUBJECT'S
"BEHAVIORAL" CUES:

Resists Custody By:
• Not Responsive to
Directions
• Evasive to Questions
• Verbal Resistance or
Body Posture
• Pulling/ Moving or
Running Away

CRIMINAL ACTIVITY
CUES:
OFFICER
MENTAL CONDITION:
OFFICER'S
ACTIONS:

CONTROL
Verbal Persuasion
EMPTY HAND TECHNIQUES

P
R
E
S
E
N
C
E
* The RCM'M must be used in cOIlJunclJon WIth Inslfucuonal
materials and only by an approved GJTe Instructor.

ASSESSMENT
• Custody Decision
• Close Distance
CONTROL BY
• Escort Position
• Distraction Techniques
• Compliance Holds
• Leveraged Takedowns
• Impact Takedowns
• Chemical

.......
.....

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Colorado Springs

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

St. Petersburg
RECO~LI\n: ND.!J)

RESPOI\"SE TO R:ESISt:.~"<cr M..crrux

R
I

s
I

s

" """".... 001

,. Aw ..",,·rhyoie>d
4A<o:moPlo,... ..

T
A
N

c

l

h.,.... PbysIcooI

1

v.-."

I
1 fr_<.

L

•
V

•
L
S

l'IIo,...

-- -I !- ---.- ---

--- -- -- --

000
00D
- I 1-0 DD
-- I IDDD
-- --I 1-- -- -1--1--1 1-- 1--1--1 1-- I 1 1 1 1 DDD
1-- 1--1--1 1-- I 1 1 1, I DDD
. - ••• r
••• •• •" !' "• •"• •I ' • ••
•,• • i' i ! t ;,
•• • •
1- 11-1-1--1-1-- 1--

- I
-- I
-- I
I

H

<

.,•
•

•
,
,

-.

~-

1

•
~

,!'•

,
•

I•

•

i•• I

~

I

t

~

••

- - -

- - -

c.-........

......

•

3

-_INC*" -5
6
4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Fort Wayne

Resistance Control Continuum:
1. Officer Presence: A display of authority by the visual pnsence of the uniform and nonverbal communicMion such as demeanor , stride, and stance.
2. Verbal Direction - Directions and commands given to the suspect(s ).
3. Soft Empty Hand Techniques - Refers to restraining and detaining of the suspect(s) with the
intent of gaining control (e.g., holding/grasping, touch pressure points, escorts, joint
manipulations and/or hlkedowns, handcuffs, leg restraints , OC spray).
Note: Distraction techniques prece(le joint manipulations and takedowns. This escalMion
of force is recognized as a necessity for the safe a nd propel' completion of the procedure.
4. Hard Empty Hand Techniques - Actiye counter measures or any technique that involves
striking the suspect(s ) (e.g., strikes, kirks, knees, forearms used for pain COlnlllhmce, stunning,
distraction, balance displacement, and motor dysfunction).
5. IntermediMe \Veapons - \Veapons that are used to assist gaining control of a situation
involving resistive, threatening or combative suspects (e.g., Impact weapons strikes and blocks,
CED, K-9, Less lethal munitions).
6. Deadly Force - Force thM creMes a substantial risk ofsel'ious bodily injll1'y or deMh.

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

KNOXVILLE POLICE DEPARTMENT
USE OF FORCE CONTINUUM

Rill tent

ValatHe

Resistant
(Passive)

Tactical
Level Two

Compliant

Comp"nce T.........
...".., Hand Scft, IIatDn Scft,
Taser, Impact Munitions, cap Stun,
Baton Hard, Empty Hcnd Hard

(Adivel

Strategic
Level One

Soft. Balon Soft
Contact Controls,

Empty Hand

Direction Controls

Officer Presence
Cooperative Controls, Verbal
Commands. Direction Controls

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Appendix C

NATIONAL SURVEY OF POLICE OFFICERS:
Examining Less than Lethal Force Policies

Conducted by:

Dr. William Terrill

MICHIGAN STATE UNIVERSITY
School of Criminal Justice
and

Dr. Eugene A. Paoline III

UNIVERSITY OF CENTRAL FLORIDA
Department of Criminal Justice

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

SECTION I – We are interested in your views regarding XXPD’s OVERALL less-than-lethal force policy.
Please indicate the extent to which you agree or disagree with the following statements, by circling the number
which best represents your opinion of the policy.

Agree
Strongly

Agree
Somewhat

Disagree
Somewhat

Disagree
Strongly

1

XXPD’s policy assists officer decision making.

1

2

3

4

2

XXPD’s policy hinders officer decision making.

1

2

3

4

3

XXPD’s policy is not restrictive enough.

1

2

3

4

4

XXPD’s policy is too restrictive.

1

2

3

4

5

XXPD’s policy is clear.

1

2

3

4

6

XXPD’s policy regarding supervisory review
of use of force reports is fair.

1

2

3

4

7

XXPD’s policy provides adequate guidance
in terms of how to apply force.

1

2

3

4

8

XXPD’s policy provides adequate guidance
in terms of when force can and cannot be used.

1

2

3

4

9

XXPD’s policy provides adequate guidance
in terms of when a force report should be completed.

1

2

3

4

SECTION II – We are interested in your views regarding SPECIFIC COMPONENTS of XXPD’s less-thanlethal force policy as it relates to dealing with resistant citizens. Please indicate the extent to which you agree
or disagree with the following statements, by circling the number which best represents your opinion of the
policy.

1

Agree
Strongly

Agree
Somewhat

1
1
1
1
1

2
2
2
2
2

Disagree
Somewhat

Disagree
Strongly

XXPD’s policy assists officer decision making
when dealing with citizens who are:
[A]
[B]
[C]
[D]
[E]

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

3
3
3
3
3

4
4
4
4
4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

2

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

1
1
1
1
1

2
2
2
2
2

3
3
3
3
3

4
4
4
4
4

XXPD’s policy is clear in terms of how to deal with
citizens who are:
[A]
[B]
[C]
[D]
[E]

6

Disagree
Strongly

XXPD’s policy is too restrictive when dealing
with citizens who are:
[A]
[B]
[C]
[D]
[E]

5

Disagree
Somewhat

XXPD’s policy is not restrictive enough
when dealing with citizens who are:
[A]
[B]
[C]
[D]
[E]

4

Agree
Somewhat

XXPD’s policy hinders officer decision making
when dealing with citizens who are:
[A]
[B]
[C]
[D]
[E]

3

Agree
Strongly

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

XXPD’s policy provides adequate guidance in terms
of how to apply force on citizens who are:
[A]
[B]
[C]
[D]
[E]

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

7

Agree
Strongly

Agree
Somewhat

1
1
1
1
1

2
2
2
2
2

Disagree
Somewhat

Disagree
Strongly

XXPD’s policy provides adequate guidance in terms
of when force can and cannot be used on citizens
who are:
[A]
[B]
[C]
[D]
[E]

non-resistant or compliant
passively resistant
verbally resistant
physically resistant in a non-assaultive nature
physically resistant in an assaultive manner

3
3
3
3
3

4
4
4
4
4

SECTION III – We are interested in your views regarding outcomes of XXPD’s less-than-lethal force policy.
Please indicate the extent to which you agree or disagree with the following statements, by circling the number
which best represents your opinion of the policy.
Increases
the likelihood

Decreases
the likelihood

Neither Increases
nor Decreases
the likelihood

1

What effect does XXPD’s policy have
on potential citizen injuries?

1

2

3

2

What effect does XXPD’s policy have
on potential officer injuries?

1

2

3

3

What effect does XXPD’s policy have
on potential citizen complaints?

1

2

3

4

What effect does XXPD’s policy have
on potential lawsuits?

1

2

3

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

SECTION IV – We are interested in your views regarding XXPD’s training with respect to less-than-lethal force.
Please indicate the extent to which you agree or disagree with the following statements, by circling the number
which best represents your opinion.

1

The pre-service Academy training that I have
received on less-than-lethal force is adequate.

Agree
Strongly

Agree
Somewhat

1

2

Disagree
Somewhat

3

Disagree
Strongly

4

1a. What, if any, part of this training should receive
more attention _______________________________________________________________
2

The In-service training that I have received on
less-than-lethal force is adequate.

1

2

3

4

2a. What, if any, part of this training should receive
more attention _______________________________________________________________
SECTION V – In terms of dealing with the following types of resistant citizens, please indicate which of the
following responses YOU believe are appropriate, irrespective of what XXPD’s policy states, by placing
a check (√) next to ALL THAT APPLY.
1

Which is/are appropriate when faced with a non-resistant/compliant citizen:
__ Verbal direction, other than commands/threats (for example – questioning)
__ Verbal commands/threats
__ Soft-empty hand techniques (for example – grabbing, shoving)
__ Pain compliance techniques (for example – pressure point control)
__ Hard-empty hand techniques (for example – striking with fists)
__ Cuffing (hands, legs)
__ Flashlight
__ Baton (for example – ASP)
__ Chemical-irritant sprays (for example – oleoresin capsicum)
__ Electronic devices (for example – TASER)
__ Projectile launchers (for example – beanbag)
__ Deadly Force

2

Which is/are appropriate when faced with a passively resistant citizen:
__ Verbal direction, other than commands/threats (for example – questioning)
__ Verbal commands/threats
__ Soft-empty hand techniques (for example – grabbing, shoving)
__ Pain compliance techniques (for example – pressure point control)
__ Hard-empty hand techniques (for example – striking with fists)
__ Cuffing (hands, legs)
__ Flashlight
__ Baton (for example – ASP)
__ Chemical-irritant sprays (for example – oleoresin capsicum)
__ Electronic devices (for example – TASER)
__ Projectile launchers (for example – beanbag)
__ Deadly Force

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

3

Which is/are appropriate when faced with a verbally resistant citizen:
__ Verbal direction, other than commands/threats (for example – questioning)
__ Verbal commands/threats
__ Soft-empty hand techniques (for example – grabbing, shoving)
__ Pain compliance techniques (for example – pressure point control)
__ Hard-empty hand techniques (for example – striking with fists)
__ Cuffing (hands, legs)
__ Flashlight
__ Baton (for example – ASP)
__ Chemical-irritant sprays (for example – oleoresin capsicum)
__ Electronic devices (for example – TASER)
__ Projectile launchers (for example – beanbag)
__ Deadly Force

4

Which is/are appropriate when faced with a physically non-assaultive citizen:
__ Verbal direction, other than commands/threats (for example – questioning)
__ Verbal commands/threats
__ Soft-empty hand techniques (for example – grabbing, shoving)
__ Pain compliance techniques (for example – pressure point control)
__ Hard-empty hand techniques (for example – striking with fists)
__ Cuffing (hands, legs)
__ Flashlight
__ Baton (for example – ASP)
__ Chemical-irritant sprays (for example – oleoresin capsicum)
__ Electronic devices (for example – TASER)
__ Projectile launchers (for example – beanbag)
__ Deadly Force

5

Which is/are appropriate when faced with a physically assaultive citizen:
__ Verbal direction, other than commands/threats (for example – questioning)
__ Verbal commands/threats
__ Soft-empty hand techniques (for example – grabbing, shoving)
__ Pain compliance techniques (for example – pressure point control)
__ Hard-empty hand techniques (for example – striking with fists)
__ Cuffing (hands, legs)
__ Flashlight
__ Baton (for example – ASP)
__ Chemical-irritant sprays (for example – oleoresin capsicum)
__ Electronic devices (for example – TASER)
__ Projectile launchers (for example – beanbag)
__ Deadly Force

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

SECTION VI – We are interested in your views regarding various components of police work. Please indicate
the extent to which you agree or disagree with the following statements, by circling the number which best
represents your opinion.
Agree
Strongly

Agree
Somewhat

Disagree
Somewhat

Disagree
Strongly

1

I work in a dangerous job.

1

2

3

4

2

A good police officer takes charge of encounters
with citizens.

1

2

3

4

3

Most people have no idea how difficult a police
officer's job is.

1

2

3

4

4

I would not consider taking another job.

1

2

3

4

5

When I started my policing career, other officers
were a valuable source of information on how to
perform as an officer.

1

2

3

4

6

There are clear, planned goals and objectives for
my job.

1

2

3

4

7

Enforcing the law is by far a patrol officer=s most
important responsibility.

1

2

3

4

8

When I first arrive on a scene or during any
interaction with a citizen, I size up the situation
in order to establish and maintain control.

1

2

3

4

9

The code of silence is an essential part of the
mutual trust necessary to good policing.

1

2

3

4

10 A good police officer is one who patrols
aggressively by stopping cars, checking out
people, running license checks, and so forth.

1

2

3

4

11 I’m usually calm and at ease when I’m working.

1

2

3

4

12 My supervisor looks out for the personal welfare
of his/her subordinates.

1

2

3

4

13 My supervisor=s approach tends to discourage
me from giving extra effort.

1

2

3

4

14 My supervisor will support me when I am wrong,
even if it makes things difficult for him or her.

1

2

3

4

15 Police officers have reason to be suspicious
of most citizens.

1

2

3

4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agree
Strongly

Agree
Somewhat

Disagree
Somewhat

Disagree
Strongly

16 In my job, a person stands a good chance
of getting hurt.

1

2

3

4

17 Police should not have to handle calls that involve
social or personal problems where no crime is
involved.

1

2

3

4

18 When an officer does a particularly good job, top
management will publicly recognize his or her
performance.

1

2

3

4

19 When an officer gets written up for minor violations
of the rules, he or she will be treated fairly by
top management.

1

2

3

4

20 When an officer contributes to a team effort rather
than look good individually top management here
will recognize it.

1

2

3

4

21 In order to remain effective, the police officer
should remain detached from the community.

1

2

3

4

22 I try to teach younger officers how to perform their
duties as an officer.

1

2

3

4

23 I know what is exactly expected of me.

1

2

3

4

24 When I’m at work I often feel tense or uptight.

1

2

3

4

25 I like my job better than the average police
officer does.

1

2

3

4

26 Given my choice, when off-duty, I would rather hang
around with non-police than other police officers.

1

2

3

4

27 Law enforcement officers should be required to do
something about - public nuisances (e.g., loud
parties, barking dogs, etc).

1

2

3

4

28 Law enforcement officers should be required to do
something about - neighbor disputes.

1

2

3

4

29 Law enforcement officers should be required to do
something about - family disputes.

1

2

3

4

30 Law enforcement officers should be required to do
something about - litter and trash.

1

2

3

4

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Agree
Strongly

Agree
Somewhat

Disagree
Somewhat

Disagree
Strongly

31 Law enforcement officers should be required to do
1
something about - parents who don’t control their kids.

2

3

4

32 Law enforcement officers should be required to do
something about - nuisance businesses that cause
lots of problems for neighbors.

1

2

3

4

33 There is a camaraderie and bond among police
officers that those outside of policing would not
understand.

1

2

3

4

34 Police officers have reason to be distrustful of
most citizens.

1

2

3

4

35 An officer is more effective when s/he patrols for
serious felony violations rather than stopping people
for minor traffic violations and misdemeanors.

1

2

3

4

36 Most law enforcement officers have to spend too
much of their time handling unimportant, non-crime
calls for service.

1

2

3

4

37 Most of what I know about policing was learned
“on the job”.

1

2

3

4

38 I know what my responsibilities are.

1

2

3

4

39 My job is a lot more dangerous than other
kinds of jobs.

1

2

3

4

40 A lot of time my job makes me very frustrated
or angry.

1

2

3

4

41 I find real enjoyment in my job.

1

2

3

4

42 When I am on the street, protecting a fellow officer
is one of my highest priorities.

1

2

3

4

SECTION VII – In this last section, we have a few background questions. Please circle your responses.
1

Did you work as a sworn officer for any other police department before XXPD?
[1] No
[2] Yes (if yes, for how many years?______________________)

2

What is your currently assigned division? _________________________
2a How long have you been assigned to your current division? _______years ________months

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

3

Within your division, please identify the response area you are assigned to or patrol most often.
Response Area: _______________
3a How long have you worked this response area? _______years ________months

4

What was your pre-service Training Academy recruit class number?____________________

5

Do you have any military experience?
[1] No
[2] Yes (if yes, do you believe that your military experience helped prepare you to be a police officer?)
1. No
2. Yes

6

Are/were any of your family member’s police officers?
[1] No
[2] Yes (if yes, please identify their relationship [e.g., father, sister, uncle, etc.] ________________)

7

Marital status:
[1] Single
[2] Married
[3] Separated
[4] Divorced
[5] Widowed

8

Would you say that, for you personally, getting promoted to a higher rank is:
[1] Very important
[2] Somewhat important
[3] Somewhat unimportant
[4] Very unimportant

9

Would you say that, for you personally, moving to a specialized unit is:
[1] Very important
[2] Somewhat important
[3] Somewhat unimportant
[4] Very unimportant

10 What is your current assignment?
[1] Patrol Officer
[2] Bikes
[3] H.I.T.S.
[4] Street Crimes Task Force

[5] School Resource Officer (SRO)
[6] K-9
[7] Other (please, identify) ________________________

11 What rank do you expect to have when you retire?
[1] Patrol Officer
[5] Deputy Chief
[2] Sergeant
[6] Chief
[3] Captain
[7] Other (please, identify) ________________________
[4] Major

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

12 What is the highest level of formal education that you completed?
[1] Less than high school (please skip to Question 13)
[2] High school diploma or GED (please skip to Question 13)
[3] Some junior college, but did not earn a degree
[4] Associates degree (AA)
[5] More than two years of college, but did not earn a bachelors degree
[6] Bachelors degree
DEGREE MAJOR___________________________
DEGREE OBTAINED ____ before police employment
____ while employed with XXPD
[7] Some graduate courses, but did not earn a graduate degree
[8] Graduate degree
DEGREE MAJOR___________________________
DEGREE OBTAINED ____ before police employment
____ while employed with XXPD
12a. Do you believe that your college experiences helped prepare you to be a police
officer?
1. No
2. Yes
13 If you were not a police officer, what do you think your occupation would be?
_________________________________________________________________________________
14 Which of the following best captures XXPD’s less-than-lethal force policy:
[1] Linear Design (for example, step, ladder, stair, PPCT, FLETC model) - often presented in the form
of a ladder or hierarchical steps of subject resistance and police force options.
[2] Wheel/Half-Wheel Design – often presented in the form of a series of concentric circles/half-circles
of subject resistance and police force options.
[3] Matrix/Box Design – often presented in the form of subject resistance and police force options
along a vertical and horizontal axis.
[4] Other (please specify) ____________________________________________________________
_________________________________________________________________________________
Finally, what (if any) part of XXPD’s less-than-lethal force policy do you think should be changed?
_______________________________________________________________________________
_______________________________________________________________________________

I.D. #_______

This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Appendix D

Assessing Police Use of Force Policy and Outcomes
In the Fall of 2005, the National Institute of Justice (NIJ) funded researchers from Michigan State
University and the University of Central Florida to conduct a study examining varying types of
nonlethal force policies. As part of this study, we wish to learn how policies are understood and
viewed by front line personnel. In particular, we are interested in your view as to the extent to
which your agency’s policy offers guidance and the ability to control suspects. Your participation
in this study will help lead to informed policy understanding and development.
We seek your cooperation in gathering this information by completing the survey, which is
expected to take 15 minutes. Your participation is voluntary and your answers are strictly
confidential. Your confidentiality will be protected to the maximum extent allowable by law as
guaranteed by federal statue (42 United States Code 3789g). You may choose not to participate at
all, refuse to participate in certain procedures or answer certain questions, or discontinue
participation at any time without penalty.
There are no known risks associated with participation in this study. Only the research team will
see your responses and the survey will be destroyed upon completion of the project at the end of
2008. The identification number at the bottom of the page is used solely to remove you from the
survey list. The information collected will be compiled, analyzed, and presented only in grouped
form with no individual agency identified.
If you have any questions about this study please contact the principal investigator, Professor
William Terrill, Michigan State University, School of Criminal Justice, 532 Baker Hall, East
Lansing, MI 48864, 517-353-9752 (phone), 517-432-1787 (fax), terrillw@msu.edu (email) or
Professor Eugene A. Paoline III, University of Central Florida, P.O. Box 161600, Orlando, Fl.
32816, 407-823-4946 (phone), 407-823-5360 (fax) epaoline@mail.ucf.edu. If you have any
questions or concerns regarding your rights as a study participant, or are dissatisfied at any time
with any aspect of this study, you may contact - anonymously, if you wish, Peter Vasilenko, Ph.D.,
Director of the Human Research Protection Programs (HRPP) at Michigan State University: (517)
355-2180, fax: (517) 432-4503, email: irb@msu.edu, or regular mail: 202 Olds Hall, East Lansing,
MI 48824, or Barbara Ward, Institutional Review Board (IRB), University of Central Florida,
12201 Research Parkway, Suite 501, Orlando, FL., 32826-3246, 407-823-2901.
You indicate your voluntary agreement to participate in this research and have your answers
included in the data set by signing below and completing the survey. Thank you for your assistance
in this effort to enhance policy understanding and development.
___________________________
Officer Signature

__________
Date

Respectfully,

_____________________
Professor William Terrill
Michigan State University

________________________
Professor Eugene A. Paoline III
University of Central Florida
I.D. #_______