What Factors Influence an Officer’s Decision to Shoot? The Promise and Limitations of Using Public Data

Date01 June 2017
AuthorStephen A. Bishopp,John L. Worrall,Scott W. Phillips,Andrew P. Wheeler
Published date01 June 2017
DOI10.1177/1525107118759900
Subject MatterResearch Articles
Research Article
What Factors Influence
an Officer’s Decision to
Shoot? The Promise and
Limitations of Using
Public Data
Andrew P. Wheeler
1
, Scott W. Phillips
2
,
John L. Worrall
1
, and Stephen A. Bishopp
3
Abstract
We analyze a set of 207 Dallas Police Department officer-involved shooting incidents
in reference to 1,702 instances in which officers from the same agency drew their
firearms but did not shoot at the suspect. We find that situational factors of whether
the suspect was armed and whether an officer was injured were the best predictors of
the decision to shoot. We also find that African Americans are less likely than Whites
to be shot. It is important to collect data on encounters in which weapons are and are
not discharged. Analyses examining only shootings is fundamentally limited in assessing
racial bias.
Keywords
officer-involved shooting, use of force, racial bias, policing
In response to high-profile shooting incidents of unarmed minority suspects in the
United States, several scholars have attempted to empirically identify factors influen-
cing police officers’ decisions whether to fire their weapons at citizens (James, James,
1
School of Economic, Political, and Policy Sciences—Criminology Program, The University of Texas at
Dallas, Richardson, TX, USA
2
SUNY Buffalo State, Buffalo, NY, USA
3
Caruth Police Institute, Dallas Police Department, Dallas, TX, USA
Corresponding Author:
Andrew P. Wheeler, School of Economic, Political, and Policy Sciences—Criminology Program, The
University of Texas at Dallas, 800 West Campbell Road, Mail Station GR 31, Richardson, TX 75080, USA.
Email: apwheele@gmail.com
Justice Research and Policy
2017, Vol. 18(1) 48-76
ªThe Author(s) 2018
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1525107118759900
journals.sagepub.com/home/jrx
& Vila, 2016; Klinger, Rosenfeld, Isom, & Deckard, 2016; Nix, Campbell, Byers, &
Alpert, 2017). This research has spawned questions regarding how often police offi-
cers shoot, who does the shooting, who is shot, and the circumstances and locations of
the shooting incidents. It has also renewed interest in the issue of whether police
officers are more likely to shoot at minority suspects relative to nonminorities for
either reasons of overt racism (Goldkamp, 1976; Sherman & Langworthy, 1979) or
implicit bias (Nix et al., 2017).
Despite recent advances in officer-involved shooting (OIS) research, the body of
work in this area is limited in several respects. First, in the United States, there is
currently no national database of OIS incidents (Alpert, 2016; Klinger, 2012) because
the data currently available from the Federal Bureau of Investigation (FBI; via the
supplementary homicide reports) are voluntarily provided by police agencies. These
data commonly include only those incidents that resulted in a fatal outcome, which
misrepresent our ability to understand the true nature of the decision to shoot (Klinger,
2012). Calls for a standardized national database requiring police agencies to provide
details regarding OIS incidents recognize the limitations of voluntary programs for
collecting quality data. For example, Alpert (2016, p. 24) stated that:
without required compliance and financial incentives, it is unlikely that the data from
many law enforcement agencies would be received and if many would be of sufficient
quality to rely on.
In response to a lack of official national data, various crowd-sourced (e.g., theguar
dian.com; fatalencounters.org) and media-driven data collections (e.g., Washington
Post data, e.g. https://www.washingtonpost.com/graphics/national/police-shootings-
2016/) have attempted to fill the gap. Such databases, though, have been criticized
for which incidents they do and do not include (Selby, Singleton, & Flosi, 2016) as
well as how different aspects of the incidents they report are coded (Klinger &
Slocum, 2017).
While recently assembled OIS databases claim national samples in the United
States, a fuller understanding of the use of deadly force should include all incidents
regardless of whether the intended target was killed, injured, or hit (Fyfe, 1978).
Whether a suspect is killed in a police shooting appears to be random with respect
to many offender and situational characteristics (Zimring, 1972), and officers often
miss their intended target (Fachner & Carter, 2015; White, 2006). In Dallas, police
officers are trained to shoot for center mass in order to “incapacitate” (Morrison &
Vila, 1998; Parent & Verdun-Jones, 1998), so they sometimes injure suspects rather
than kill them.
A more fundamental limitation of available OIS data is that they fail to take into
account instances in which officers chose not to use deadly force (Reiss, 1980). This is
an important consideration in OIS research, as prior work suggests that most police
officers do not use deadly force even when it seems justified (Pinizzotto, Davis,
Bohrer, & Infanti, 2012). To estimate the probability a suspect will, conditional on
their race, be shot, it is necessary to have a set of control cases in which officers had
Wheeler et al. 49

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT