Research Article

T a s e r s i n t h e M e d i a

Examining fatal and nonfatal

incidents involving the TASER
Identifying predictors of suspect death reported in the media

Michael D. White
Justin Ready
Arizona State University

Research Summary
According to TASER International, nearly 10,000 police departments in the United States
have deployed the TASER as a less lethal force alternative in some capacity. Despite the
TASER’s increasing popularity, serious questions have been raised about the device’s physi-
ological side effects; in particular, Amnesty International has reported that more than 300
people have died after being subjected to the TASER. Although a growing body of research
has examined the physiological effects of the TASER on animals and healthy human vol-
unteers in laboratory settings, there has been virtually no empirical analysis of “real-world”
fatal and nonfatal TASER cases simultaneously. This article examines all media reports of
TASER incidents from 2002 to 2006 through a comprehensive review of LexisNexis and
New York Times archives. We compare TASER incidents in which a fatality occurred to
TASER incidents in which a fatality did not occur and then employ multivariate analyses
to identify the incident and suspect characteristics that are predictive of articles describing
TASER-proximate deaths.

Policy Implications
Several suspect factors were significantly associated with the reporting of a fatal TASER
incident, including drug use (but not alcohol), mental illness, and continued resistance.
Multiple deployments of the TASER against a suspect was also associated with the likelihood
of the article describing a fatality—especially if the suspect was emotionally disturbed—
which raises the possibility that the risk of multiple shocks might not be uniform for all
suspects. More research is needed to explore the relationship between mental illness, drug
use (illicit or therapeutic), continued resistance, and increased risk of death. In the mean-
time, police departments should develop specific policies and training governing the use of

Volume 8 • Issue 4 863

R esearch Ar ticle TASERS in the M edia

multiple TASER shocks against individuals who could be in these vulnerable physiological
and psychological states.

Keywords
TASER, conducted energy device, police use of force, less lethal weapons, media

I
n testimony before the U.S. Department of Justice inquiry into “deaths in custody fol-
lowing electro-muscular disruption,” representatives from Amnesty International (2007:
4) stated:

AI has collected data on more than 290 cases of individuals in the United States
and Canada who since 2001 have died after being struck by police Tasers. While
in most cases medical examiners have attributed death to other factors…AI has
identified at least 20 cases where coroners have found the TASER served a causal
or contributory factor in the death and other cases where the TASER was cited as
a possible factor in autopsy reports.

In July 2008, a grand jury in Louisiana indicted a police officer on a manslaughter charge
involving the death of a suspect who was shocked nine times with a TASER. During the grand
jury proceeding, the District Attorney said, “It is our intention to show at trial that Mr. Nugent
[the officer] caused the death of Baron Pikes by Tasing him multiple times, unnecessarily and
in violation of Louisiana law” (“Officer charged in death of Tasered man,” 2008). Alterna-
tively, at the conclusion of the Department of Justice inquiry on deaths after electromuscular
disruption, the National Institute of Justice published a special report stating that “there is no
conclusive medical evidence within the state of current research that indicates a high risk of
serious injury or death from the direct effects of CED [TASER] exposure” (National Institute
of Justice, 2008: 3).1
These recent events illustrate the ongoing controversy surrounding the physiological effect
of the TASER. Although available evidence suggests that deaths after TASER deployment (i.e.,
TASER-proximate deaths) are rare, there is little understanding of why fatalities occur in a small
number of cases and how—if at all—the TASER contributed to those deaths.2 Unfortunately,
researchers’ ability to explore causes of TASER-proximate deaths has been constrained by two
data limitations. First, this emerging technology is not used in a vacuum and laboratory stud-
ies on the physiological effects of the device cannot control for circumstances that develop in

1. Conducted energy device (CED) is the general term used to describe the TASER and other versions of the
device. Even though TASER dominates the CED market with approximately 95% of sales in the United
States, competitors include Stinger Systems and Law Enforcement Associates. Stinger Systems has sold
12,000 weapons since 2000. Law Enforcement Associates introduced their CED in March 2005.
2. “TASER-proximate” death is the term the authors use to describe an incident where the suspect was struck
with a TASER and subsequently died, but no definitive evidence exists to identify the TASER as the cause of
death.

864 Criminology & Public Policy

 White, R eady

a natural setting, such as the presence of drugs and alcohol, poor health of suspects, physical
confrontations, the use of handcuffs and other less lethal weapons, and elevated levels of adrena-
line and physical exertion. Amnesty International (2007: 4) recognized the contributing role
of these real-world circumstances, explaining “most of the individuals who died were agitated,
disturbed and under the influence of drugs, and/or had underlying health problems such as
heart disease.”3 The degree to which the TASER—when combined with one or more of these
circumstances—increases the risk of death remains unknown.
Second, empirical research to date has failed to study fatal and nonfatal TASER incidents
simultaneously. Although studies involving coroners’ reports examined only those cases in
which suspects died, laboratory research and studies using police use-of-force data focused on
deployments in which most suspects survived. Neither approach can produce variation in the
outcome that is required to develop predictive models with advanced statistical methods—that
is, to identify predictors of suspect death. As a result of these two limitations, critical questions
about fatal and nonfatal incidents remain unanswered. For example: Do TASER incidents that
result in death differ in notable ways from nonfatal incidents? If so, what are the distinguishing
characteristics of TASER deployments that result in death? And how do specific risk factors
alone—or in combination with other circumstances—increase the likelihood of suspect death?
These questions are cause for concern in communities and police departments across the United
States because the TASER has replaced chemical sprays and impact weapons as the preferred
less lethal alternative for U.S. police.
This article explores these questions by analyzing data collected from all media reports
describing police use of the TASER from 2002 to 2006 (N = 521), which were generated
through a broad search of LexisNexis and the New York Times. We used media reports because
they are the only data source that provides information about the circumstances surrounding
fatal and nonfatal TASER deployments. First, we used bivariate analyses to compare news reports
that described fatal deployments (n = 188) with articles describing nonfatal deployments (n =
333). Then we used multivariate analyses, including logistic regression and CHAID (a form
of segmentation modeling) to identify suspect and incident characteristics that are significant
predictors of news reports involving TASER deaths. The article concludes with a discussion
of implications for police policy, training, and practice, as well as the ongoing debate about
police use of the TASER.

Prior Research
The Context: Police and the Use of Force
Police officers have legal authority to use force in a wide range of situations, including empty
hand or physical force, less lethal weapons (i.e., baton, pepper spray, or CED) and, as a last resort,
the use of a firearm (Walker and Katz, 2002). Despite its central role in policing (e.g., Bittner,

3. In the Louisiana case cited above, the coroner’s report found that Pikes was handcuffed and lying on the
ground when the police deployed the TASER. He was also physically exhausted after running from the
police and had a history of cocaine involvement.

Volume 8 • Issue 4 865

R esearch Ar ticle TASERS in the M edia

1970), research indicates that police use of force is statistically rare, occurring in approximately
1% of police–citizen encounters (Bureau of Justice Statistics, 1999).4 That said, because there
are approximately 43 million police–citizen encounters in a given year, an estimated 421,000
use-of-force incidents occur annually, which translates into approximately 1,100 incidents in
a typical day.
The application of physical force may potentially have devastating consequences, not only
for the suspect and the officer but also for the law-enforcement agency, the community, and
police–community relations (Fyfe, 1988; Geller and Scott, 1992). For example, the National
Advisory Commission on Civil Disorders (1968) concluded that the police were a cause of
many riots occurring during the 1960s. More recent examples of police–citizen encounters
that triggered social turbulence include the L.A. riots after the acquittal of the four officers
who were videotaped beating Rodney King in 1991, the strained community relations in New
York City after the shooting of Amadou Diallo in 1999, and the civil unrest after the Sean Bell
incident in 2006.
Because of the physical and social harms—as well as the political fallout that may result
from abuses of force—police agencies have expanded their alternatives to firearms. The Presi-
dent’s Commission on Law Enforcement and the Administration of Justice made this central
to the national policing agenda in 1965 when it proposed developing a range of less lethal
weapons. Throughout several decades, advances in technology have led to the development of
force alternatives such as oleoresin capsicum (OC; i.e., pepper) spray, impact weapons, foams,
ballistic rounds, nets and, most recently, conducted energy devices (CEDs; Wrobeleski and
Hess, 2003). This expanding arsenal is intended to provide officers with a broader range of
options when a situation requires the application of force but has not escalated to the point
when deadly force is necessary, which therefore reduces the likelihood of serious injury and
death to officers and suspects.
During the 1990s, OC spray became popular among police agencies in the United States,
and this trend serves as the backdrop for the current work on CEDs because many of the same
issues and concerns have been raised (Smith and Alpert, 2000). Specifically, controversies sur-
rounding police use of OC spray have centered upon its use on passive resisters, disproportionate
use on racial minorities, and potential health risks (Kaminski, Edwards, and Johnson, 1999).
Several studies have examined the effectiveness of OC spray and concluded that the weapon
effectively incapacitates the suspect in most deployments, results in fewer officer injuries, and
reduces police reliance on other types of force (Gauvin, 1994; Kaminski, Edwards, and Johnson,
1998; Lumb and Friday, 1997; Nowicki, 1993).

4. This estimate becomes much greater when handcuffing and verbal commands are included as use of
force.

866 Criminology & Public Policy

 White, R eady

The TASER
CEDs are becoming a visible component of U.S. policing and media depictions of police
behavior. The most popular CED is the TASER (manufactured by TASER International,
Scottsdale, AZ), and recent estimates indicate that more than 10,000 law-enforcement agen-
cies have purchased and deployed TASERs in some capacity (most commonly, the M26 and
X26 models). The TASER fires two probes at a rate of 180 feet per second, and when striking
the subject, delivers a 50,000-volt shock during a 5-second cycle (Vilke and Chan, 2007).
Although the technology has the capacity to cause acute discomfort, it is not designed to be a
pain compliance weapon:

CEDs work by incapacitating volitional control of the body. These weapons create
intense involuntary contractions of skeletal muscle, causing subjects to lose the
ability to directly control the actions of their voluntary muscles. CEDs directly
stimulate motor nerve and muscle tissue, overriding the central nervous system
control and causing incapacitation regardless.… This effect terminates as soon as
the electrical discharge is halted. Immediately after the TASER shock, subjects are
usually able to perform at their physical baseline (Vilke and Chan, 2007: 349).

Researchers have not kept pace with the rapid growth of this technology and, until recently,
most of the available information about the TASER came from the CED industry, internal
police reports, and publications from civil and human rights organizations such as Amnesty
International and the American Civil Liberties Union. As adoption of the technology has spread,
however, questions have emerged in three general areas: (1) policy (When and against whom
is it appropriate to use the TASER?), (2) effectiveness (Does the device work as intended?),
and (3) physiological impact (Does the TASER increase the risk of injury or death?). These
issues are discussed below, but greater emphasis is placed on physiological effect because it is
the focus of this article.
When is it appropriate to use the TASER? Police agencies provide guidance to officers on
the appropriate use of force through a force continuum that describes the verbal and physical
actions a police officer can take in response to different levels of suspect resistance and behavior.
Police departments have varied considerably in terms of where they place the TASER on the
force continuum, particularly in the earlier years when the device first became available. A central
issue is whether the TASER should be used on suspects who are not following verbal commands
or who are passively resisting police efforts. This issue became highly controversial when the
media captured images of two separate incidents in which the police used the TASER on a
college student who was being passively resistant. The first incident occurred in a University of
California at Los Angeles library in November 2006 and the second took place during a press
conference for Senator John Kerry at the University of Florida in 2007. Other questionable
police practices spotlighted in the media include use of the device on vulnerable populations,
such as the physically and mentally disabled, pregnant women, children, and the elderly. The

Volume 8 • Issue 4 867

R esearch Ar ticle TASERS in the M edia

Police Executive Research Forum (PERF, 2005) and the International Association of Chiefs of
Police (IACP, 2005) have issued policy recommendations to offer guidance to agencies in their
use of CEDs. Both the IACP and PERF suggest that CEDs be used against only those who
are actively resisting, that they not be used against minors or the elderly except in emergency
situations, and that each deployment be closely supervised.
Does the TASER work effectively? Research examining the effectiveness of the device in the
field has focused on two questions: (1) Does the TASER have the intended physiological effect,
thereby terminating suspect resistance; and more generally, (2) does use of the device reduce
suspect and officer injuries? Unfortunately, limited empirical research is available to answer
these questions. With regard to suspect resistance, field data analyzed by TASER International
(2006) and internal evaluations by police agencies (Seattle Police Department, 2004) place the
effectiveness rate of the TASER somewhere between 80% and 94%. White and Ready (2007)
analyzed TASER deployment records from the New York City Police Department and found
that suspects stopped resisting and were successfully incapacitated in 86% of incidents occurring
during a 4-year period. In a follow-up study, White and Ready (2009) found that the effect
of the device on suspect resistance was mitigated by several factors, including suspect weight,
intoxication, and the distance between the suspect and the officer.
Several police agencies that have implemented CEDs on a broad scale have later reported
reductions in injuries sustained during police–citizen contacts. Police departments in Austin,
Texas; Putnam County, Florida; and Cincinnati, Ohio, reported reductions in injuries to
suspects and officers after adopting the TASER (Jenkinson, Neeson, and Bleetman, 2006;
TASER International, 2006). Smith, Kaminski, Rojek, Alpert, and Mathis (2007) examined
injuries to suspects and officers in two departments and concluded that use of CEDs was as-
sociated with reduced odds of injury in one department but not the other. The authors noted
that “not every agency’s experience will be the same regarding CED use and injuries” (Smith
et al., 2007: 439).
Does the TASER increase the likelihood of suspect death? Amnesty International has
spearheaded efforts to raise awareness of the potentially harmful physiological effects of the
TASER. Concern about the possible health risks associated with CEDs has prompted a grow-
ing body of research involving reviews of coroner reports in death cases, comprehensive reviews
of empirical research, and biomedical research using animals and healthy human volunteers.
Kornblum and Reddy (1991) examined 16 deaths after TASER deployments and determined
that drug overdose was the cause of death in most incidents.5 Ordog, Wasserberger, Schlater,
and Balasubramanium (1987) examined 218 emergency room cases that involved suspects
who were shocked with a TASER; three suspects died and all had PCP in their systems. Strote,
Campbell, Pease, Hamman, and Hutson (2006) examined autopsy reports in 28 TASER-related
deaths and found that the device was not identified as the direct cause of death in any of the

5. In one of the cases, the authors concluded that the TASER could have contributed to the suspect’s death
because he had a history of cardiac disease, although the suspect also had lethal levels of PCP in his
system (see also Vilke and Chan, 2007).

868 Criminology & Public Policy

 White, R eady

cases; however, the authors noted that it was listed as a potential contributing factor in 21%
of the fatalities (see also Canadian Police Research Centre, 2005; Joint Non-Lethal Weapons
Human Effects Center of Excellence, 2005).
Several studies have examined the physiological impact of the TASER on animals (e.g., pigs
and dogs) and healthy human volunteers, focusing specifically on cardiac rhythm disturbances
such as ventricular fibrillation (e.g., McDaniel, Stratbucker, and Smith, 2000; McDaniel,
Stratbucker, Nerheim, and Brewer, 2005; Roy and Podgorski, 1989; Stratbucker, Roeder, and
Nerheim, 2003). For instance, both Levine, Sloane, Chan, Vilke, and Dunford (2005) and
Ho et al. (2006) monitored heart function in human volunteers who were subjected to the
TASER and found no evidence of changes in heart rhythm or functioning. In their review of
this research, Vilke and Chan (2007: 353) concluded the following:

The potential for life-threatening cardiac dysrhythmias or cardiac muscle damage

to occur as a result of the electrical discharge from current TASER devices appears
to be low based on the available studies. However, there may be theoretical risks to
patients with pacemakers or underlying cardiac disease, and the effect of recurrent
or prolonged TASER discharges remains unclear.

Vilke and Chan (2007: 353) noted that existing research has yet to investigate “non-cardiac
effects” of the TASER, including the device’s effect on metabolism (i.e., potassium, sodium, and
pH levels in the blood) and respiration (i.e., carbon dioxide elimination and respiratory rate).
Laboratory studies using animals and healthy human volunteers represent an important
step in understanding the physiological effect of the TASER, but the implications of this research
may be limited by fundamental differences between the test subjects and those individuals
most likely to experience the TASER in a real-world setting. That is, suspects who are typically
shocked by the TASER during police–citizen encounters are less likely to be healthy, they are
not voluntary participants, they are frequently under the influence of drugs or alcohol, they
often have preexisting physical or mental conditions, and many are physically resistant during
the TASER deployment. Any of these conditions occurring in a natural setting can lead to an
increased risk of physiological harm. In October 2007, Wake Forest University Medical Center
released their findings from a large-scale evaluation that involved physician reviews of 1,000
real-world TASER deployments, thereby avoiding many limitations associated with laboratory
research using animals and human volunteers. The Wake Forest research team coordinated with
six law-enforcement agencies. When an officer in one of those agencies deployed the TASER,
the suspect was transported to a hospital for evaluation by a research team doctor. The study
found that 99.7% of suspects had minor or no injuries.6 William Bozeman, the study’s lead
investigator, reported:

The injury rate is low and most injuries appear to be minor. These results support
the safety of the devices (Wake Forest University Baptist Medical Center, 2007).

6. Two suspects died, but neither death occurred as a result of the TASER.

Volume 8 • Issue 4 869

R esearch Ar ticle TASERS in the M edia

Summary
Although police frequently rely on the TASER when they experience resistance from crime
suspects and disorderly individuals, many questions are unanswered concerning the physiological
effects of the device when deployed in a natural setting. Given the absence of a causal link to
suspect fatalities, why do deaths occur in a small number of incidents? Also, are there identifi-
able risk factors involving the behavior of the suspect or the circumstances of the encounter
that increase the likelihood of death? If so, what are the implications of those risk factors for
police policy and training? The following analysis begins to explore these issues using national
media data describing fatal and nonfatal TASER incidents.

Methodology
We conducted a broad search of media reports via LexisNexis and New York Times Select us-
ing keyword searches to identify all articles involving the TASER from January 2002 through
December 2006.7 We relied on media data because such data are the only viable data source
with detailed information on fatal and nonfatal cases. After the universe of news reports was
identified (N = 691), we recorded information for 68 variables relating to the content of the
articles and the circumstances under which police used the TASER, placing special emphasis on
characteristics of the suspect, officer, and events leading to the deployment.8 Several categories
of cases were excluded from the analysis. First, our study focused only on cases where a police
officer deployed the TASER. Two types of news stories did not satisfy this criterion: business
reports for TASER International (i.e., stock reports) and articles describing general trends or
research relating to police use of the TASER. Additionally, a small number of news reports
(approximately 30; 5.4%) were eliminated because of missing information. These stories were
typically one or two sentences long and provided little detail about the incident. The authors
set aside these reports because their inclusion added little value to the descriptive or multivariate
analyses that are the central focus of the article. Finally, duplicate cases were excluded to prevent
certain incidents from being overrepresented in the analysis. Duplicate cases were defined as
any news report that contained the same information as another report that described a specific
incident in which police used the TASER on a suspect (often published on the same date).9 In
all, 170 articles were excluded from the analysis.
The content analysis was based on 521 nonduplicate articles, each of which described an
incident in which police used the TASER on a suspect. The dependent variable in this study

7. We used a range of keywords to conduct the database searches. These terms included “TASER,”“CED,” and
“electronic stun device.”
8. We developed our coding instrument based on a TASER deployment reporting form used by a major met-
ropolitan police department. Also, the instrument was focused on types of information typically included
in articles about the TASER. Unfortunately, certain types of information such as the race, height, and
weight of the suspects; the distance to the suspect; and many officer characteristics could not be included
in the final analysis because a large proportion of the news reports did not contain this information.
9. The duplicate cases were identified by cross-referencing the articles using the date of the incident, city,
suspect’s name, and the newspaper in which the article was published.

870 Criminology & Public Policy

 White, R eady

was whether the article described a TASER incident resulting in death—yes or no. In all, 36%
of the articles (n = 188) described an incident in which the suspect died after being shocked
with a TASER and the remaining 64% (n = 333) of the articles describe nonfatal TASER in-
cidents. Articles that described TASER-proximate deaths were overrepresented in the data, as
research indicates that deaths after TASER use are rare (Wake Forest University Baptist Medical
Center, 2007).

Limitations and Considerations of the Data Source

Three types of potential bias are associated with use of media data on TASER events, and each
warrants some discussion. The first potential bias stems from differences in the “newsworthiness”
of fatal and nonfatal incidents involving the TASER. That is, nonfatal incidents are less likely to
be reported in the news because they lack the inherent “drama” associated with fatal incidents.
Although the extent of this bias remains unknown, two issues might mitigate its impact in this
study. First, given the high level of interest and debate surrounding the device, it could be that
nearly all TASER incidents—even nonfatal cases—have the same likelihood of being reported
in the media and therefore being captured in our data.10 Second, media reporting of nonfatal
incidents could be most likely when some other interesting characteristic is associated with the
event, such as repeated use of the device against a suspect or use against vulnerable populations
(e.g., the mentally ill and elderly). In effect, our data-collection approach likely captured similar
types of TASER events—those with newsworthy characteristics—that had different outcomes—
fatality or no fatality. In short, these nonfatal-though-newsworthy events may actually represent
the best comparison for fatal TASER incidents.
The second potential bias concerns how the measures were operationalized and the quality
of information recorded from the media reports. We recognized this limitation from the outset
and attempted to address it in several ways. First, police reports and interviews of police officers
are the primary source of information used by news reporters for most crime and justice issues,
and use of the TASER is no exception. In this sense, concerns regarding quality in media data
mirror concerns with official police data and other data sources. Second, we examined the extent
of missing data across fatal and nonfatal articles and found notable consistencies, especially
among some of the key variables. For example, the number of times the TASER was deployed
during the incident was reported in nearly all news stories (i.e., this information was missing
in 4% of nonfatal incidents and 3% of fatal incidents). Alternatively, officer and suspect race
were almost never reported, regardless of incident outcome (i.e., in 1% in both types of articles).
Articles that described fatal and nonfatal incidents were also similar in their reporting of suspect
mental illness (missing in 22% and 28%, respectively) and drug use (missing in 27% and 33%,
respectively). At a minimum, similarities in missing data suggest some degree of consistency in
data quality across article type (fatal and nonfatal incidents).

10. Although the extent of news coverage may vary based on the outcome (fatal or nonfatal), our data collec-
tion approach captures all those media-reported incidents.

Volume 8 • Issue 4 871

R esearch Ar ticle TASERS in the M edia

Last, each question in the content analysis instrument was designed so that the coder
recorded whether a specific piece of information was reported in the article. The coders were
not allowed to speculate about any characteristics of the police-citizen encounters.11 Moreover,
in cases in which multiple articles were published about the same event, we compared stories to
corroborate “facts” and identify discrepancies.12 As a result, the main source of error was missing
information that could result in a false negative—such as when a coder correctly indicated that
an article did not report that the suspect was mentally ill when in fact he was. We attempted
to minimize this type of error by focusing the data collection and analyses on items that were
expected to yield reasonably accurate data.13
The third potential concern is source bias; that is, media representations of police use of
the TASER could differ from actual TASER use. Some circumstances could mitigate concern
over this bias. First, currently little research is available on police use of the TASER, particu-
larly studies comparing fatal and nonfatal incidents (although the NIJ funded a six-site study
in 2008–2009 and other studies are likely underway). Second, although an extensive body of
literature has documented media bias in depicting police practices (Chermak, 1995; Surette,
1998), research also suggests that the police are much less effective in “controlling the media
message” when the use of force is involved (Chermak, McGarrell, and Gruenewald, 2006; Tuch
and Weitzer, 1997). Third, Ready, White, and Fisher (2008: 27) compared news reports with
police records on TASER deployments and found “noteworthy consistencies across data sources
with regard to many suspect and incident-related characteristics.”
In short, this study reflects an effort to identify predictors of TASER-related deaths as
reported in the news media. Although the actual influences of the biases described above is
unknown, we believe that the article makes a substantive contribution to this underdeveloped
area of research.14

Analysis
We first compared articles describing fatal and nonfatal TASER incidents on the bivariate level
across various suspect and incident characteristics. We then used two multivariate approaches—
logistic regression and CHAID (a form of segmentation modeling) to identify significant
predictors of whether an article describes a TASER-related death. Binary logistic regression was

11. The content analysis was completed by us and six graduate students from the John Jay College of
Criminal Justice. For example, one item in the instrument asks whether the article described the suspect as
mentally ill or emotionally disturbed (yes/no). The question did not measure whether it seemed that the
suspect was mentally ill based on the narrative describing the encounter. The researchers were instructed
to only check “yes” if the article explicitly described the subject as possessing this trait. Thus, the unit of
analysis is the article itself, rather than the encounter.
12. In cases where two articles provided inconsistency or contradictory information (which was infrequent),
the authors coded information from the article that provided the most detail.
13. The authors also compared these items with available data on the TASER from one police department as a
measure of convergent validity.
14. The authors are grateful to Senior Editor Steve Chermak for assistance in delineating the strengths and
weaknesses of the data source.

872 Criminology & Public Policy

 White, R eady

employed because the dependent variable is dichotomous with a “yes” or “no” response. Similar
to logistic regression, CHAID predicts the probability of an event occurring, but the method
relies on different assumptions and properties and uses segmentation modeling to achieve the
task. CHAID divides a population into increasingly homogenous segments that differ based on
the dependent variable—in this case, whether the article describes a TASER-proximate death
(Jones, Harris, Fader, and Grubstein 2001). The resulting segments are mutually exclusive and
exhaustive. As the analysis proceeds, the best predictor is selected among a particular subgroup
of cases based on chi-square analyses.
CHAID analysis was employed in this study because it offers several practical advantages.
First, CHAID is useful for identifying different sets of predictors for subsets of a population
(Jones et al., 2001). For example, predictors of a fatal outcome may be different for intoxicated
and sober suspects, and CHAID facilitates the identification and exploration of these interac-
tions. Second, the use of multiple methods allows researchers to “triangulate” their findings or to
identify inconsistencies across techniques. Third, CHAID has been used in the study of police
practices previously, including the use of force and arrest decisions (Kane, 1999; White, 2002,
2006, 2008). Finally, an added benefit of CHAID is the user-friendly visual representation of
complex analyses and relationships for practitioners and policymakers.

Findings
Table 1 provides a frequency distribution for the basic content of all LexisNexis and New York
Times articles about the TASER from January 2002 through December 2006. LexisNexis articles
account for 84.8% (n = 586) of the cases, whereas the remaining 15.2% (n = 105) of the cases
were derived from the New York Times. Overall, 75.4% of the news reports describe a specific
incident in which the police used the TASER on a suspect. The volume of news publications
about the TASER increased from 24 (3.5%) in 2002 to 179 (25.9%) in 2004, and then peaked
at 338 articles (48.9%) in 2006. The regions of the country that were the source of the most
articles are the Southeast (n = 248) and the West (n = 190). The Northeast has produced the
fewest articles relating to the TASER, accounting for 8.4% (n = 58) of cases.15 As a whole, the
news reports offered considerable detail about the circumstances under which officers used the
device in the field, the events leading up to the deployment, and the behavior and characteristics
of the suspects and officers. However, a relatively small proportion of the articles also discussed
when it is appropriate for police officers to use the weapon (17.4%) and the physiological effects
(or effectiveness) of the TASER (18.0%).

Ta b l e 1

15. Many articles published in the New York Times were not counted in the Northeast because they
duplicated LexisNexis articles that originated in other regions of the country. The states that generated
the most news coverage of the TASER are Florida (n = 122), California (n = 77), and Colorado (n = 60; not
shown in Table 1). Altogether, these three states accounted for 37.5% of non-duplicate news reports about
the TASER during the study period.

Volume 8 • Issue 4 873

R esearch Ar ticle TASERS in the M edia

Content Analysis of LexisNexis and New York Times

Articles about the TASER, 2002–2006
Variable Percent Valid Cases

Source of the article

LexisNexis 84.8 586
New York Times 15.2 105
Total 100.0 691
Type of Story
Duplicate case or not a TASER deployment* 24.6 170
Total 100.0 691
Did the TASER deployment result in death?
Yes 36.1 188
No 63.9 333
Total 100.0 521
Did the article discuss when it is appropriate for police to use the TASER?
Yes 17.4 120
No 82.6 571
Total 100.0 691
Did the article discuss the physiological effects (or effectiveness) of the TASER?
Yes 18.0 124
No 82.0 567
Total 100.0 691
Year of publication
2002 3.5 24
2003 6.4 44
2004 25.9 179
2005 15.3 106
2006 48.9 338
Total 100.0 691
Region of country
Northeast 8.4 58
Southeast 35.9 248
Midwest 18.8 130
West 27.5 190
Not in United States/Not applicable 9.4 65
Total 100.0 691

Comparative Findings
Table 2 provides a comparison of news reports describing fatal and nonfatal TASER deployments
during the 5-year study period (2002–2006). The variables in the analysis were separated into
two categories: incident and suspect characteristics. Several differences are noteworthy. The
articles describing TASER incidents that resulted in death were more likely to involve situa-
tions in which the police officer used the weapon more than once on the same suspect (50.8%
vs. 23.3% for nonfatal cases), suspect resistance continued after TASER deployment (38.8%
vs. 22.7% for nonfatal), the suspect was handcuffed (i.e., in custody) when the TASER was
deployed (22.3% vs. 6.1% for nonfatal cases), and the suspect was transported to the hospital

874 Criminology & Public Policy

 White, R eady

(58.6% vs. 29.8% for nonfatal cases).16 Table 3 shows the relationship between the number
of times the TASER is used on the same suspect and a fatal outcome in the news media. Ap-
proximately one fourth of the articles that describe incidents where the TASER was used once
resulted in suspect death. In the articles where the TASER was used repeatedly on the same
suspect, fatal outcomes increased considerably (50–55% for 2 or 3 deployments; 60% for 4
or more deployments).

Ta b l e 2

Content Analysis of News Reports Describing Police TASER use: A

Comparison of Fatal and Nonfatal TASER Deployments, 2002–2006
Nonfatal Fatal
Variables Percentage Valid Cases Percentage Valid Cases

Incident characteristics
(i.e., circumstances)
Officer used TASER more than once 23.3% 322 50.8% 181
Suspect continued to resist 22.7% 326 38.8% 178
Other less lethal weapon used before 16.2% 327 22.5% 182
Other less lethal weapon used after 26.1% 326 31.0% 184
Suspect in custody when TASER used 6.1% 330 22.3% 188
Suspect taken to hospital 29.8% 332 58.6% 186
Backup officer(s) present at the scene 71.9% 331 72.2% 180
Suspect characteristics
Minor or senior citizen 9.6% 332 4.3% 188
Male suspect 91.9% 332 97.9% 188
Intoxicated from alcohol 10.5% 333 8.0% 187
Under influence of drugs 6.3% 332 23.7% 186
Emotionally disturbed or mentally ill 22.9% 332 36.2% 188
Armed with weapon 37.0% 332 16.5% 188
Assaulted an officer 29.7% 330 37.1% 186
Verbal or passive resistance 26.3% 327 22.2% 185
News reports describing fatal and nonfatal TASER incidents also differed in terms of the
characteristics of suspects (see Table 2).17 Suspects in articles that described fatal incidents were
more often under the influence of drugs (23.7% vs. 6.3%) and mentally ill or emotionally
disturbed (36.2% vs. 22.9%), but they were less likely to be armed with a weapon (16.5% vs.
37.0%). A word of caution is warranted regarding the drug and mental illness variables. Police
reports and police officer statements are the primary source of information used by newspaper
reporters to document these suspect characteristics. As a result, these variables likely reflect the
police officer’s assessment of the suspect at the time of the incident based on his or her per-

16. Although approximately one third of the articles mentioned that the suspect had attacked the officer,
whether or not the officer was injured (and the extent of that injury) rarely was described.
17. Nine percent of the articles identified the suspect’s race or ethnicity (n = 51). Of those 51 articles, 16 in-
volved black suspects, 21 involved Hispanic suspects, and 13 involved white suspects. The failure to report
race and ethnicity did not extend to other demographic characteristics, as three quarters of the articles
identified the suspect’s gender and more than 60% discussed the suspect’s age.

Volume 8 • Issue 4 875

R esearch Ar ticle TASERS in the M edia

sonal observations and on information collected at the scene (e.g., from family members and
witnesses).18 This information is not based on more definitive tests such as a urinalysis, blood or
hair analysis, or clinical assessments, and the reader should bear this in mind. The next section
explores whether these incident and suspect characteristics were significant predictors of TASER
fatalities in the media when controlling for other factors through multivariate analysis.

Ta b l e 3

Content Analysis of LexisNexis and New York Times Articles,

2002–2006: Number of Times Police Used the TASER by
Suspect Death (Number of Valid Cases in Parentheses)
Number of Times Police Used the TASER Suspect Death
on a Suspect (i.e., number of cycles) No Yes

Once 73.5% 26.5%

(247) (89)
Twice 45.4% 54.6%
(44) (53)
Three times 50.0% 50.0%
(15) (15)
Four or more times 40.0% 60.0%
(16) (24)
Total 64.0% 36.0%
(322) (181)

18. Interestingly, a large portion of the articles did mention the type of drug the suspect was abusing, typically
cocaine or methamphetamine.

876 Criminology & Public Policy

 White, R eady

Multivariate Findings
Table 4 displays the results from a logistic regression model predicting TASER-related deaths
in the news media using suspect and incident circumstances as covariates. The likelihood ratio
test for the analysis was significant and the estimate for Nagelkerke R-squared indicates that
the model accounts for 33.7% of the variation in the outcome variable.

Ta b l e 4

Logistic Regression Model Predicting Police TASER

Deployments Resulting in Suspect Death
Variables B S.E. Exp(B)

Incident characteristics
Number of times TASER used 0.018 .038 1.018
Suspect continued to resist after TASER 0.820 .274 2.270**
Other less lethal weapon used before –0.028 .320 0.972
Other less lethal weapon used after 0.139 .271 1.149
Suspect in custody when TASER used 1.195 .358 3.302**
Suspect taken to hospital 1.217 .250 3.379***
Backup officer(s) present at the scene –0.429 .272 0.651
Suspect characteristics
Age of suspect (minor or senior citizen) –0.491 .484 0.612
Gender of suspect 1.679 .698 5.361*
Intoxicated from alcohol –0.892 .433 0.410*
Under influence of drugs 1.414 .351 4.112***
Emotionally disturbed or mentally ill 0.562 .261 1.754*
Armed with a weapon –1.083 .323 0.338***
Level of resistance before TASER (none or verbal)
Passive physical resistance 0.134 .347 1.143
Active physical –0.306 .318 0.736
Active potentially lethal –0.458 .442 0.632
Constant –2.740 .740 0.065
Log likelihood 477.423
R2 (Nagelkerke) .337
Chi square 131.794
d.f. 16.000
Significance .000
N 470.000

Notes. S.E. = standard error.

*p < .05. **p < .01. ***p < .001.

Volume 8 • Issue 4 877

R esearch Ar ticle TASERS in the M edia

Predictors of articles describing TASER deployments resulting in suspect death include the
following:
• Suspect resistance after the weapon was deployed (the likelihood of death was two times
greater)
• Suspect was handcuffed (i.e., in custody) when the weapon was deployed (suspect death
was more than three times as likely)
• Suspect was transported to the hospital after the TASER was deployed (the likelihood of
death was more than three times greater)
• Suspect was under the influence of drugs (suspect death was four times more likely)
• Suspect was emotionally disturbed or mentally ill (suspect death was nearly twice as
likely)19
Two factors were associated with a reduced likelihood of the article describing an incident
resulting in death: the suspect being under the influence of alcohol and the suspect being armed
with a weapon (in more than half of those articles, the weapon was a knife or other cutting
instrument; 23% were armed with a gun, however). Several variables that were not statisti-
cally significant are noteworthy, including the number of times the TASER was used on the
suspect, the use of other less lethal weapons, the suspect’s age, and the level of resistance before
the TASER was deployed.
The CHAID findings are displayed in Figure 1. The top cell (or root node) in the CHAID
tree includes all 521 articles and highlights the 36.1% in which the suspect died after being
shocked with the TASER. The initial split was made based on whether the suspect described
in the article was under the influence of drugs (hereafter referred to as “high”; this classification
does not include alcohol), thus separating the 521 TASER articles into two cells as follows: those
in which the suspect was not high (n = 453; 86.9% of the total) and those in which the suspect
was high (n = 68; 13.1% of the total). The splits in CHAID are made according to differences
in the dependent variable: In articles with high suspects, death occurred 67.7% of the time,
compared with death occurring 31.4% of the time in articles with suspects who were not high.
Another split was made from the not-high cell based on the number of times the TASER was
used on a suspect: In articles in which the suspect was not high and the TASER was deployed
repeatedly, death occurred 52.9% of the time; compared with suspect death occurring in 22.3%
of the articles in which the TASER was deployed once on suspects who were not high.
The next split was made from the cell indicating that the TASER was used more than once.
This split is based on whether the suspect was mentally ill or emotionally disturbed: In articles
in which the suspect was not high during the encounter but was emotionally disturbed, and
in which the TASER was used more than once, suspect fatalities occurred 74.6% of the time,
compared with suspect death occurring in 36.0% of the articles in which the suspect was not
emotionally disturbed or high (and the TASER was used more than once). The final split is made

19. Suspect gender was statistically significant (p > .05), but there were 31 articles describing a female suspect,
and 4 involved a TASER-proximate death. We also ran the model without the suspect gender variable and
the results were unchanged.

878 Criminology & Public Policy

 F I G U RE 1

CHAID Analysis with TASER Articles

Did suspect die after being stunned by TASER?

Node 0
Category % n
■ No 63 .92 333
■ Yes 36 .08 188
Total (100 .00) 521

Suspect described as under influence of illicit substance

Adj. p-value+0.0000, Chi-square=33.7806, df=1

NO YES, <missing>

Node 3 Node 4
Category % n Category % n
■ No 68 .65 311 ■ No 32 .35 22
■ Yes 31 .35 142 ■ Yes 67 .65 46
Total (86 .95) 453 Total (13 .05) 68

About how many time was TASER used?

Adj. p-value+0.0000, Chi-square=41.4013, df=1

1, <missing> 2 or more

Node 5 Node 6
Category % n Category % n
■ No 77 .74 248 ■ No 47 .01 63
■ Yes 22 .26 71 ■ Yes 52 .99 71
Total (61 .23) 319 Total (25 .72) 134

Article discussed suspect taken to hospital after incident Suspect described as mentally ill or emotionally disturbed
Adj. p-value=0.0167, Chi-square=7.6833, df=1 Adj. p-value=0.0000, Chi-square=19.7269, df=1

NO YES, <missing> NO YES

Node 9 Node 10 Node 7 Node 8

Category % n Category % n Category % n Category % n
■ No 82 .52 170 ■ No 69 .03 78 ■ No 64 .00 48 ■ No 25 .42 15
■ Yes 17 .48 36 ■ Yes 30 .97 35 ■ Yes 36 .00 27 ■ Yes 74 .58 44
Total (39 .54) 206 Total (21 .69) 113 Total (14 .40) 75 Total (11 .32) 59

Volume 8 • Issue 4 879

R esearch Ar ticle TASERS in the M edia

from the cell indicating that the TASER was used only once, based on whether the suspect was
transported to the hospital or not. In articles in which the suspect was not high and was shocked
by the TASER once but was taken to the hospital, suspect death occurred 30.9% of the time,
compared with death occurring in 17.5% of the articles in which the suspect was not high, the
TASER was deployed only once, and the suspect was not transported to the hospital.
Table 5 provides a summary of the termination cells for the CHAID tree, which features
the predictors, cell size, percentage of the total cases, and percentage of the dependent variable.
Five termination cells are listed, with the percentage of articles describing suspect death rang-
ing from 17.5% to 74.6%. The CHAID termination cells call attention to a relatively large
group of articles (40% of the total) with infrequent suspect deaths (17.5% describing a death),
a similar-sized group in which death occurred in one third of the articles (the second and third
subgroup combined), and two groups of articles in which suspect death is very common, oc-
curring in 68–75% of the articles. The high-risk groups of articles are characterized by suspects
who were under the influence of drugs or emotionally disturbed, and they were subjected to
more than one TASER deployment.

T ABLE 5

Summary of CHAID Analysis Termination Groups

Article describing a TASER- Number of % of % Describing
proximate death indicates… Articles Total Death

1. Suspect not under influence of illicit drug; TASER 206 39.5% 17.5%
deployed once; suspect not taken to hospital
2. Suspect not under influence of illicit drug; 113 21.7% 31.0%
TASER deployed once; suspect taken to hospital
3. Suspect not under influence of illicit drug; 75 14.4% 36.0%
TASER deployed more than once; suspect not emotionally disturbed
4. Suspect under the influence of illicit substance 68 13.1% 67.6%
5. Suspect not under influence of illicit drug; 59 11.3% 74.6%
TASER deployed more than once; suspect is emotionally disturbed

Discussion
Notable Nonsignificant Characteristics
It might be useful to explore the implications of the findings by first considering several presum-
ably important characteristics that did not emerge as statistically significant.20 One controversy
surrounding the TASER involves use of the device on vulnerable populations, notably the
young and the elderly. Arguably, the physiological makeup of both the young and elderly could
elevate the risk of potential negative health effects following TASER deployment. That is, the
elderly are more likely to suffer from physical ailments—particularly heart problems—whereas

20. These characteristics are notable either because of prior research on TASER use or because conventional
wisdom suggests they might be important.

880 Criminology & Public Policy

 White, R eady

increased risks for minors often stem from their low body weight and immature physical de-
velopment. The findings indicate, however, that the suspect’s age—measured many different
ways—was not predictive of an article describing a TASER-related death. This finding may be
a result of the small number of cases that involve those vulnerable populations because only 40
articles described an incident involving a minor or elderly person, and approximately 4% of
the TASER-proximate death articles involved such an individual.
Second, some attention has focused on whether the risk of suspect death increases when
the TASER is used in conjunction with other less lethal alternatives, such as pepper spray or the
baton, often because one of the devices was ineffective in subduing the suspect. The use of other
less lethal devices was included as three separate dichotomous variables (run independently), and
although nearly 40% of the news reports described the use of another less lethal device, none
of the variables was statistically significant. Last, there has been some disagreement about the
appropriate placement of the TASER on the force continuum and, more specifically, whether
the device should be used in response to passive resistance by a suspect. Approximately one
quarter of the news reports described a case that involved a suspect who gave only verbal or
passive resistance, but the variable was not predictive of an article describing a fatality.

Significant Predictors
The most important predictor of an article describing a TASER-proximate death was the suspect
being the under the influence of drugs. In the logistic regression, when the suspect was high, the
article was four times as likely to describe a fatality. In the CHAID model, the “suspect high”
variable served as the first split and isolated an end group in which more than two thirds of the
articles involved TASER-proximate deaths. This finding is consistent with prior research examin-
ing coroners’ reports that has documented a link between drug use, TASER deployments, and
increased risk of death. Several possible explanations for this relationship are given here. First,
despite its recurrent theme in the research, it is conceivable that the relationship between drug
use and suspect death is spurious. In many cases, a suspect under the influence of illegal drugs
is shocked with the TASER by police and death does not occur. It is possible that some other
intervening variable, that often coincides with drug use, increases the risk of death.
Second, the relationship between drug use and increased risk of death could be related to
the individual’s behavior following the TASER deployment. Continued resistance by the suspect
after TASER deployment also emerged as a significant predictor of articles describing fatalities.
Suspects who are high might be more likely to continue resisting after the TASER is deployed.
This continued physical exertion, which is a result of impaired judgment, could place enough
stress on the body that—when combined with the effects of the drug—induces severe physi-
ological responses such as excited delirium. Excited delirium (ED) syndrome is defined as:

The sudden death of an individual, during or following an episode of excited de-

lirium, in which an autopsy fails to reveal evidence of sufficient trauma or natural
disease to explain the death. In virtually all such cases, the episode of excited delirium

Volume 8 • Issue 4 881

R esearch Ar ticle TASERS in the M edia

[defined as a transient disturbance in consciousness or cognition involving violent

behavior] is terminated by a struggle with police or medical personnel, and the use
of physical restraint (DiMaio and DiMaio, 2006 :1).

Since 1984, more than 130 cases of fatal, cocaine-associated ED have been reported in the medi-
cal and forensic literature, and nearly all have involved use of force (usually by police, although
some involve medical staff at psychiatric hospitals; Sztajnkrycer and Baez, 2008). DiMaio and
DiMaio (2006: 4) explained that most deaths attributed to ED involve cocaine or metham-
phetamine abuse and that death is not caused by police use of force; rather, death is likely a
product of, “normal physiological reactions of the body to stress gone awry, and to the use of
stimulants.” In their review of 74 TASER-proximate death cases, DiMaio and DiMaio (2006:
42) concluded that these deaths “almost invariably describe individuals in excited delirium, high
on illegal stimulants who die not at the time they are ‘Tasered’ but sometime after.”
A third possible explanation for the drug-use finding involves the combined effect of the
illicit drugs and the TASER shock to the suspect’s system. Illegal drugs have a wide range of
potentially harmful effects on the human body, which include the functioning of major organs
such as the brain and heart, the central nervous system, respiration, circulation, and metabolism.
These negative effects—which by themselves can be fatal—might become more lethal when
combined with the effects of one or more shocks from the TASER. Also, the combined effect
of the TASER and an illicit drug may vary by the type of drug used. For example, results from
this study suggest that alcohol intoxication has an inverse relationship with the probability of an
article describing a TASER-proximate death. Perhaps the risk of death is greater with stimulants,
such as methamphetamines and cocaine (by far, the most common type of drugs described in
the news reports), than with depressants such as alcohol.
Mental or emotional distress was also a significant predictor in the multivariate models.
When the suspect was described as being mentally ill and in crisis, there was a greater prob-
ability that the article involved a TASER-proximate death. In particular, the CHAID analysis
isolated a small group of articles (n = 59; 10% of the articles examined) in which the suspect
was emotionally disturbed or mentally ill and the officers used the TASER more than once.
Three quarters of these articles involved a TASER-proximate death. Much like suspects who
are under the influence of drugs, those who are mentally ill and in crisis might be less likely to
acquiesce to police authority (i.e., continued resistance) because of their emotional state.21 Con-
tinued physical resistance might then prompt the police to apply more cycles from the TASER.
The number of times the device is used on a suspect has received considerable attention from
civil liberties and human rights organizations because of the potential for excessive force (i.e.,
Amnesty International indicated that the device could be used to torture a suspect) and because
of the risk of injury or death. Vilke and Chan (2007: 353) concluded, “the effect of recurrent
or prolonged taser discharges remains unclear.” The findings here raise the possibility that the
impact of multiple shocks might not be consistent for all suspects and that the risk of death

21. Recall that continued resistance by itself was a significant predictor in the logistic regression analysis.

882 Criminology & Public Policy

 White, R eady

might be greater for certain subgroups, such as those who are mentally ill. Moreover, suspects
who are mentally ill could also have psychotropic drugs in their systems, which increases the
risk of death. DiMaio and DiMaio (2006) noted that therapeutic drugs, such as antidepressants
and antipsychotics often prescribed to the mentally ill, have been linked to ED.
Several other predictors were associated with a greater likelihood that an article described
a fatality. First, if the suspect was transported to the hospital, then there was an increased likeli-
hood that the article described a TASER-proximate death. This finding makes intuitive sense, as
suspects who later died would have exhibited symptoms at the scene that led to their transport
to the hospital.22 Second, nearly one quarter of the articles describing a TASER-proximate death
involved a suspect in custody. Most of these suspects were handcuffed or being detained in a
holding cell. The nature of this relationship is unclear, particularly because in-custody deaths
are rare. A recent Bureau of Justice Statistics (2007) study documented a total of 2,002 arrest-
related (or in-custody) deaths in the United States from 2003 to 2005. Given that police in
the U.S. make approximately 1.3 million arrests each year (Bureau of Justice Statistics, 2005),
the number of in-custody deaths during that 3-year period represents less than 0.5% of arrests.
Nevertheless, when a suspect dies in police custody, the case generates substantial controversy and
often raises questions about excessive force and police brutality. The finding here may simply be
a consequence of the “newsworthiness” of in-custody deaths. However, it is worth considering
whether use of the device with handcuffed suspects is a health risk in cases in which the suspect’s
air passages are already restricted (e.g., when the individual is lying on his stomach).
Third, the suspect being armed with a weapon is inversely related to the likelihood that
the article describes an incident resulting in death. This finding too may be a consequence of
media reporting but in a slightly different way. Although much of the news reporting can be
characterized as negative or critical of the TASER, some articles were decidedly more “pro-
TASER.”23 In fact, 35 articles describe incidents in which the suspect was armed with a gun
(and 79 described suspects armed with knives or other cutting instruments). The emergence
of the “suspect armed” variable may be an artifact of this second type of news report, which
includes articles that describe “successful” TASER deployments involving armed and dangerous
suspects who were thwarted by the less lethal device.

Implications for Policing

The implications of this study for police departments center on four issues. First, substantial
variation exists across departments with regard to policy, use, and reporting of TASER deploy-
ments. Variation in critical issues—such as when the TASER should be used and who it should
be used against—has resulted in controversial incidents and could be related to an increased
likelihood of fatal outcomes. Although the limited scope of research on the TASER prohibits

22. Some police departments, such as the New York City Police Department, require that all suspects who are
shocked with a TASER be transported to the hospital for examination. The recent guidelines by PERF and
IACP make similar recommendations.
23. We make this “anti-TASER” assertion based on the content analysis of more than 500 news articles.

Volume 8 • Issue 4 883

R esearch Ar ticle TASERS in the M edia

any sort of discussion related to “best practices,” the PERF and the International Association
of Chiefs of Police (IACP) offer model policy guidelines involving TASER use. In the absence
of “best practices,” we believe departments would be well advised to consult these guidelines
carefully when crafting their own policies. Adherence to PERF/IACP standards regarding suspect
resistance level, use against vulnerable persons, multiple deployments, and reporting practices
will likely reduce the potential for controversial and inappropriate deployments.
Second, police departments should routinely collect and analyze data on their officers’ TASER
use, and they should consider making that data publicly available. With regard to the first point,
many police departments require that officers file reports when any degree of force is used, but
not all departments do. Kane (2007: 775) noted that the primary motivations for collecting data
on use of force is to identify problem officers and he explained that to ensure that “police depart-
ments that collect data on their practices can identify far better than others policies that work,
policies that do not work, and areas of organizational behavior that should be regulated.” The
second point is more contentious—that police should make those data available to the public.
Although police are traditionally reluctant to release data, especially sensitive information such as
use-of-force data, there are two reasons for public release of TASER data. First, there is a critical
need for transparency in this area. Use of force in general, and use of the TASER specifically,
can cause long-term damage to a department’s relationship with the community. By releasing
data to the community, the department can convey a powerful message of accountability and
openness. Moreover, Kane argued that police have an obligation to share such data because the
public funds and “owns” all information generated by the department:

Currently, however, police departments operate under a paradigm where the burden
of proof rests on the members of the public to justify adequately why they want
access to police data. The paradigm should change such that the burden of proof
is on police departments to justify why the information should be kept unavail-
able to the public. Until this paradigm shifts, police accountability in the United
States will remain stalled, and members of the public will continue to be harmed
unnecessarily by the police (Kane, 2007: 779).

Third, police departments should seek partnerships with local researchers to study use and
deployment of the TASER. Departments could continue to explore some of the intuitive and
counterintuitive findings from this study. In particular, departments could explore the potential
differential effects of the TASER with certain types of drugs. Is the device more or less effective on
suspects under the influence of cocaine, methamphetamine, or alcohol? Are injuries or medical
emergencies more common with certain types of drugs? And how does physical exertion and
active resistance interact with these substances? Partnerships between police and local researchers
could also explore tactical issues such as multiple deployments, the use of multiple less-lethal
weapons, and the so-called “velcro effect”—where simply showing the device may produce

884 Criminology & Public Policy

 White, R eady

suspect compliance.24 Voluntary release of data by police, coupled with analysis by independent
researchers, would move knowledge in the field forward and represent a significant step toward
transparency with possibly profound implications for police–community relations.
Fourth, police departments should become familiar with the medical research on the ef-
fects of the TASER and excited delirium. Officers typically receive training on how to use the
TASER in the academy and during in-service “refresher” courses. Information pertaining to
medical risks, physiological responses, and high-risk medical situations should be incorporated
into these training sessions. In other words, training should move beyond the “tasing” of vol-
unteer police recruits while their fellow classmates look on in amusement—a traditional and
still popular method for introducing recruits to the TASER. Clearly, the focus of this training
should remain on operational and tactical issues, but trainers should also provide officers with
state-of-the-art medical information so that, once on the street, officers are keenly aware of the
potential risks and can quickly identify medical emergencies as they occur.

Research and Policy Implications

There are also some important next steps for researchers and funding agencies. First, additional
studies should explore the value of media data for studying police field behavior, especially use
of force. One must be wary of many biases; nevertheless, research has failed to document the
actual extent or pervasiveness of bias. If concerns over these biases can be properly addressed,
this methodological approach could then be extended to the study of other types of police
field behavior where data are traditionally difficult to obtain (e.g., automobile pursuits and
deadly force).
Second, researchers and government agencies would benefit from a continued dialogue on
the creation of a national use-of-force reporting system, which would include both lethal and less
lethal force incidents (e.g., TASER deployments). Hickman, Piquero, and Garner (2008) made
important strides in this area by highlighting the shortcomings of the Police–Public Contact
Survey, and policy essays by Klinger (2008) and Smith (2008) offered thoughtful guidance on
how best to initiate and structure such a system.25 Clearly, our understanding of physiological,
tactical, and policy issues associated with TASER use would expand immeasurably through
analysis of incidents in a national reporting system.
Third, the call for more research in this area coincides with the increased availability of
funding through the Economic Recovery Act of 2009. Although the NIJ continues to be a vi-
able source of funding for police research (with a track record in supporting TASER research),
social scientists should seek opportunities in nontraditional venues, such as the National Science
Foundation, National Institutes of Health, and the Centers for Disease Control and Preven-
tion. The study of CEDs generally, and the TASER specifically, has been fragmented often

24. We thank one of the anonymous reviewers for introducing us to this term.
25. Hickman et al. (2008) also noted that other nations such as New Zealand have already developed national
use-of-force reporting systems that the United States could model.

Volume 8 • Issue 4 885

R esearch Ar ticle TASERS in the M edia

with little interaction among the relevant fields. Partnerships among social scientists, physi-
cians (or others in the medical field), pharmacologists, psychiatrists, and police practitioners
could broaden this line of research through empirical studies that are both comprehensive and
multidisciplinary. Such an approach would be ideal for examining the perceived relationships
between mental illness, substance abuse, active resistance, multiple TASER deployments, and
an increased risk of death.

References
Amnesty International. 2007. Amnesty International’s concerns about TASER use:
Statement to the U.S. Justice Department inquiry into deaths in custody. London, UK:
Amnesty International.
Bittner, Egon. 1970. The functions of police in modern society. Washington, DC: U.S.
Government Printing Office.
Bureau of Justice Statistics. 1999. Contacts between police and the public: Findings from the
1999 national survey. Washington, DC: U.S. Department of Justice.
Bureau of Justice Statistics. 2005. Contacts between police and the public: Findings from the
2002 national survey. Washington, DC: U.S. Department of Justice.
Bureau of Justice Statistics. 2007. Arrest-related deaths in the United States, 2003–2005.
Washington, DC: U.S. Department of Justice.
Canadian Police Research Centre. 2005. Review of conducted energy devices. Ottawa, Ontario:
Author.
Chermak, Steven. 1995. Image control: How police affect the presentation of crime news.
American Journal of Police, XIV: 21–43.
Chermak, Steven, Edmund McGarrell, and Jeff Gruenewald. 2006. Media coverage of police
misconduct and attitudes toward police. Policing: An International Journal of Police
Strategies and Management, 29: 261–281.
DiMaio, Theresa G. and Vincent J.M. DiMaio. 2006. Excited delirium syndrome: Cause of
death and prevention. New York: Taylor and Francis.
Fyfe, James J. 1988. Police use of deadly force: Research and reform. Justice Quarterly, 5:
165–205.
Gauvin, Roy. 1994. Oleoresin capsicum spray: A progress report. Portland, OR: Portland
Police Department.
Geller, William and Michael S. Scott. 1992. Deadly force: What we know. Washington, DC:
Police Executive Research Forum.
Hickman, Matthew J., Alex R. Piquero, and Joel H. Garner. 2008. Toward a national
estimate of police use of nonlethal force. Criminology & Public Policy, 7: 563–604.
Ho, Jeffrey D., James R. Miner, Dhamunjaya Lakireddy, Laura L. Bultman, Laura Heegaard,
and William G. Heegaard. 2006. Cardiovascular and physiologic effects of conducted
electrical weapon discharge in resting adults. Academic Emergency Medicine, 6:
239–243.
International Association of Chiefs of Police (IACP). 2005. Electro-muscular disruption
technology (EMDT): A nine-step strategy for effective deployment. Alexandria, VA: Author.

886 Criminology & Public Policy

 White, R eady

Jenkinson, Emma, Clara Neeson, and Anthony Bleetman. 2006. The relative risk of police
use-of-force options: Evaluating the potential for deployment of electronic weaponry.
Journal of Clinical Forensic Medicine, 13: 229–241.
Joint Non-Lethal Weapons Human Effects Center of Excellence (HECOE). 2005. Human
effectiveness and risk characterization of the electromuscular incapacitation device—A
limited analysis of the TASER. Brooks Air Force Base, TX: Air Force Research
Laboratory.
Jones, Peter, Phillip Harris, Jamie Fader, and Lorie Grubstein. 2001. Identifying chronic
juvenile offenders. Justice Quarterly, 18: 479–507.
Kaminksi, Robert J., Steven M. Edwards, and James W. Johnson. 1998. The deterrent effects
of oleoresin capsicum on assaults against police: Testing the velcro-effect hypothesis.
Police Quarterly, 1: 1–20.
Kaminksi, Robert J., Steven M. Edwards, and James 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: 7–29.
Kane, Robert J. 1999. Patterns of arrest in domestic violence encounters: Identifying a police
decision-making model. Journal of Criminal Justice, 27: 65–79.
Kane, Robert J. 2007. Collect and release data on coercive police actions. Criminology &
Public Policy, 6: 773–780.
Klinger, David A. 2008. On the importance of sound measures of forceful police actions.
Criminology & Public Policy, 7: 605–617.
Kornblum, Ronald N. and Sara K. Reddy. 1991. Effects of the Taser in fatalities involving
police confrontation. Journal of Forensic Science, 36: 434–448.
Levine, Saul D., Christian M. Sloane, Theodore C. Chan, Gary M. Vilke, and James
V. Dunford. 2005. Cardiac monitoring of subjects exposed to the Taser. American
Emergency Medicine, 13: S47.
Lumb, Richard C. and Paul C. Friday. 1997. Impact of pepper spray availability on police
officer use of force decisions. Policing: An International Journal of Police Strategies and
Management, 20: 173–185.
McDaniel, Wayne C., Robert A. Stratbucker, Max Nerheim, and James E. Brewer. 2005.
Cardiac safety of neuromuscular incapacitating defense devices. Pacing and Clinical
Electrophysiology, 28: 284–287.
McDaniel, Wayne C., Robert A. Stratbucker, and Rick Smith. 2000. Surface application of
Taser stun guns does not cause ventricular fibrillation in canines. Proceedings of the Annual
International Conference of the IEEE EMBS, Chicago, IL.
National Advisory Commission on Civil Disorder (Kerner Commission). 1968. Report of the
National Advisory Commission on Civil Disorder. Washington, DC: U.S. Government
Printing Office.
National Institute of Justice. 2008. Study of deaths following electromuscular disruption: Interim
report. Washington, DC: Office of Justice Programs, U.S. Department of Justice.
Nowicki, Edward. 1993. Oleoresin capsicum: A non-lethal force alternative. Law
Enforcement Technology, 20: 24–27.

Volume 8 • Issue 4 887

R esearch Ar ticle TASERS in the M edia

“Officer charged in death of Tasered man.” 2008. CNN, August 13. Retrieved December 22,
2008 from cnn.com/2008/CRIME/08/13/la.taser.death/index.htm.
Ordog, Gary, Jonathan Wasserberger, Theodore Schlater, and Subramanium
Balasubramanium. 1987. Electronic gun (taser) injuries. Annals of Emergency Medicine,
16: 73–78.
Police Executive Research Forum (PERF). 2005. PERF conducted energy device policy and
training guidelines for consideration. Washington, DC: PERF Center on Force and
Accountability.
Ready, Justin, Michael D. White, and Christopher F. Fisher. 2008. Shock value: A
comparative analysis of news reports and official police records on TASER deployments.
Policing: An International Journal of Police Strategies and Management, 31: 148–170.
Roy, O.Z. and Andrew Podgorski. 1989. Tests on a shocking device—The stun gun. Medical
and Biological Engineering and Computing, 27: 445–448.
Seattle Police Department. 2004. Seattle Police Department TASER use and deployment fact
sheet. Retrieved November 11, 2008 from taser.com/documents/seattlepduse.pdf.
Smith, Michael R. 2008. Toward a national use-of-force collection system: One small and
focused step is better than a giant leap. Criminology & Public Policy, 7: 619–627.
Smith, Michael R. and Geoffrey P. Alpert. 2000. Pepper spray: A safe and reasonable
response to suspect verbal resistance. Policing: An International Journal of Police Strategies
and Management, 23: 233–245.
Smith, Michael R., Robert J. Kaminski, Jeffrey Rojek, Geoffrey P. Alpert, and Jason 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.
Stratbucker, Robert, Richard Roeder, and Max Nerheim. 2003. Cardiac safety of high voltage
Taser X26 waveform. Proceedings of the 25th Annual International Conference of the
IEEE EMBS, Cancun, Mexico.
Strote, Jared, Rich Campbell, John Pease, Shane Hamman, and Range Hutson. 2006. The
role of Tasers in police restraint-related deaths. Annals of Emergency Medicine, 46: S85.
Surette, Ray. 1998. Media, crime, and criminal justice: Images and realities. Belmont, CA:
Wadsworth.
Sztajnkrycer, Matthew D. and Amado A. Baez. 2008. Excited delirium and sudden unexpected
death. Retrieved January 9, 2008 from acep.org.
TASER International. 2006. TASER non-lethal weapons: Field data as of July 2006.
Retrieved September 23, 2006 from taser.com/facts/documents/Injury_Reduction_
Stats_2Q_2006.ppt.
Tuch, Steven A. and Ronald Weitzer. 1997. The polls-trends: Racial differences in attitudes
toward the police. Public Opinion Quarterly, 61: 642–663.
Vilke, Gary M. and Theodore C. Chan. 2007. Less lethal technology: Medical issues.
Policing: An International Journal of Police Strategies and Management, 30: 341–357.

888 Criminology & Public Policy

 White, R eady

Wake Forest University Baptist Medical Center. 2007. Press release: Nationwide independent
Taser study results suggest devices are safe. Retrieved October 27, 2007 from 1.wfubmc.ed/
news/articleID=2165.
Walker, Samuel and Charles M. Katz. 2002. The police in America: An introduction. New
York: McGraw Hill.
White, Michael D. 2002. Identifying situational predictors of police shootings using
multivariate analysis. Policing: An International Journal of Police Strategies and
Management, 25: 726–751.
White, Michael D. 2006. Hitting the target (or not): Comparing characteristics of fatal,
injurious, and non-injurious police shootings. Police Quarterly, 9: 303–330.
White, Michael D. 2008. Identifying good cops early: Predicting recruit performance in the
academy. Police Quarterly, 11: 27–49.
White, Michael D. and Justin Ready. 2007. The TASER as a less-lethal force alternative:
Findings on use and effectiveness in a large metropolitan police agency. Police Quarterly,
10: 170–191.
White, Michael D. and Justin Ready. 2009. The impact of the TASER on suspect resistance:
Identifying predictors of effectiveness. Crime & Delinquency. Pre-published in
OnlineFirst February 26, 2008 DOI 10.1177/0011128707308099.
Wrobeleski, Henry M. and Karen M. Hess. 2003. Police operations. Belmont, CA:
Wadsworth.

Michael D. White is an associate professor in the School of Criminology and Criminal Justice
at Arizona State University. White’s primary research interests involve the police, including
use of force, misconduct, training, and performance measurement. His recent work has been
published in Criminal Justice and Behavior, Journal of Experimental Criminology, and Crime &
Delinquency.

Justin Ready is an assistant professor in the School of Criminology and Criminal Justice at
Arizona State University. He has conducted research on crime displacement, hot-spot policing
strategies, victimology, and surveillance cameras. His articles have appeared in Justice Quarterly,
Crime & Delinquency, Criminology, and Police Quarterly.

Volume 8 • Issue 4 889