An Assessment of Officer-Involved Shooting Data Transparency in the United States

Abstract

Although researchers and the public intuitively know that the lack of available data on the topic of officer-involved shootings (OIS) is a problem, the scope of the problem has not been identified. A lack of transparency regarding OIS data contributes to the legitimacy crisis facing policing in the United States. Valid and reliable OIS data would create the ability to craft and assess informed public policy. In addition, police organizations’ roles could then be evaluated allowing for accountability and additional training based on OIS incidents. Employing content analysis of websites for police agencies, their municipal/county governments, and sheriffs’ offices servicing populations of 100,000 or more according to the 2010 U.S. Census, the availability of OIS data is identified and classified. It was found that 155 (of 277) police agencies and 96 (of 548) sheriffs’ offices made some form of data available related to OIS incidents across five categories (annual reports, OIS summary reports, individualized shooting reports, press releases, and spreadsheet reports). The current assessment is the first in the literature to systematically evaluate the transparency of law enforcement organizations servicing large jurisdictions regarding OIS incidents. Those reports and data available are described and placed within the context of the recommended national database proposed by Klinger and colleagues to begin to assess the extent of OIS data transparency issues facing policing. In addition, agencies providing data were compared utilizing one-way analysis of variance on a number of structural variables drawn from the 2008 Census of State and Local Law Enforcement Agencies to evaluate any patterns demonstrated by those organizations providing data. Findings suggest great variation in the type, classification, and quality of data presented by law enforcement organizations, which limits its utility for the purposes of research as well as policy creation and evaluation.

Keywords

officer-involved shooting lethal force deadly use of force police data data transparency

Although researchers and the public intuitively know that the lack of available data on the topic of officer-involved shootings (OIS) is a problem, the scope of the problem has not been identified. The current assessment of police agency transparency explores the extent and classification of OIS reporting for municipal and sheriffs’ agencies serving jurisdictions of 100,000 or more residents. A lack of transparency regarding OIS has contributed to the current legitimacy crisis facing law enforcement in the United States (Chanin & Espinosa, 2016). The field of policing has experienced legitimacy crises in the past; however, the current crisis related to police use of deadly force, specifically OIS, appears to hold an increased level of sustainability. Although the causes of its sustainability are open to interpretation and debate, a consistent theme in research related to use of force is the recommendation for increased transparency through public access to law enforcement records (Alpert, 2016; Fyfe, 1981, 1982, 1988, 2002; Kane, 2007; Robin, 1963). What is less clear in these calls for additional data collection and ease of access to such data is the current state of publicly available OIS law enforcement data.

Research related to use of deadly force by the police is not a new phenomenon (see Robin, 1963), but historically there has been no systematic data collection effort by governmental agencies related to police use of force (Fyfe, 2002). The fact that an accurate account of the number of citizens killed by police in the United States was not available even surprised former Federal Bureau of Investigation (FBI) Director James Comey (Williams et al., 2019). As early as 1978, Fyfe recommended the collection of data regardless of outcome when police fire their weapons at citizens. Following several recent high-profile OIS incidents, researchers have suggested that changes related to this issue may be possible (White, 2016; Williams et al., 2019). These cases may have indeed spurred the federal government forward as the FBI (2019) recently began an optional national data collection effort related to police use of force, to include OIS data. This data collection effort is a partnership between the law enforcement community and the FBI, and it will record use of force incidents resulting in serious bodily injury or death to include incidents where an officer’s firearm is discharged at a person (FBI, 2019). Similar to the suggestions of Klinger and colleagues (2016), the FBI data collection effort will classify information according to incident, subject, and officer (FBI, 2019). As noted above, however, the law enforcement community will submit use of force information to the FBI on a voluntary basis.

The lack of publicly available OIS data is not a new discovery. Research, however, has not progressed beyond the identification of the problem. Researchers must move beyond acknowledging the lack of OIS data and begin to search for the boundaries of this issue. What types of data are available? How are those data presented to the public? For those agencies that provide data, is it consistent across agencies (e.g., method of presentation, variables presented)? The current assessment of available data related to OIS incidents seeks to take those next steps in addressing the scope of the OIS data problem. Through an analysis of police agency, municipal/county government, and sheriffs’ offices’ websites, a number of agencies that provide a broad range of data related to OIS incidents within their jurisdictions were identified. Building upon prior literature, this research defines OIS incidents as all events in which an officer/sheriff’s deputy discharges their firearm in their role as a law enforcement official (Fyfe, 1980, 1982; Klinger et al., 2016; Sparger & Giacopassi, 1992; White, 2001).

Valid and reliable OIS data create the ability to make informed public policy decisions. Without OIS data, researchers across many fields have not been able to fully assess the societal impact of OIS. At present, much of the broader dialogue surrounding OIS is driven by activists, social media, and media reporting. This narrative has become politicized suggesting that police participate in systemic misconduct, motivated by racial bias, and officers involved are presumed guilty (Weitzer, 2015). OIS data would allow researchers to evaluate OIS over time to better understand (a) if the politicized narrative is accurate; (b) how to formulate policy to address the findings of any empirical research; (c) whether public policies have been effective; and (d) how to hold police accountable for their role in OIS.

In an effort to assess the consistency of data collected and their reporting, available data are assessed within the framework of Klinger et al.’s (2016) recommended national database on police use of deadly force. This framework builds upon ideal lists of OIS variables that have been defined and refined by researchers since the 1970s (Fyfe, 1978; Kane, 2007; Klinger et al., 2016; Reiss, 1971). Klinger et al. (2016) presented five broad areas that should serve as the minimum for data collection:

. . . (a) the number of incidents in which police officers discharge firearms at citizens; (b) the demographic characteristics of the officers and citizens involved in each incident; (c) the agency/agencies employing the involved officers and location of each incident; (d) the particular weapon(s) used by police and citizens; and (e) the injuries, if any, suffered by officers and citizens. (p. 215)

The FBI (2019) recently began data collection efforts that largely mirror those suggested by Klinger and colleagues (2016) creating a link between the academic literature and law enforcement practice. Therefore, researchers must take the next step in this exploratory process by identifying and assessing early adopters of OIS data collection and dissemination efforts. By evaluating the proportion of adopters versus nonadopters, researchers can begin to evaluate the quantitative extent of OIS data transparency issues facing law enforcement. Now that the FBI has adopted measures, the field has begun the process of institutionalization and progressed beyond the search for an ideal grouping of OIS variables. The literature assessing OIS data availability is briefly discussed below. We intend only to provide an overview of this extensive literature rather than providing comprehensive coverage. We then describe the breadth of our search and findings related to the current state of law enforcement transparency in providing OIS data to the public.

Literature Review

According to Kane (2007), police in a democratic society should be accountable to the public that they serve, which should have the ability to audit the police. Public access to OIS data would allow for the formulation of evidence-based public policy and evaluation of public policy, and create the ability to hold law enforcement accountable when necessary. The police, however, typically do not collect and maintain comprehensive records related to their uses of force, and, when they do, it is considered proprietary information not for public distribution (Fyfe, 1988; Kane, 2007). This lack of transparency is evidenced by the state of extant literature addressing use of deadly force. Although this literature spans more than 50 years (see Robin, 1963), calls for uniform data collection date back almost as far (Fyfe, 1978). Yet, these calls for uniform data collection and distribution have largely gone unrealized. This is why much of the use of deadly force literature draws on data from individual jurisdictions open to individual researchers. Frequently, these data obtained from individual jurisdictions have been related to incidents involving the discharge of police firearms. Shooting data have been analyzed related to police agencies in New York City (Fyfe, 1980), Memphis (Fyfe, 1982; Sparger & Giacopassi, 1992), Philadelphia (White, 2001, 2002, 2006), and St. Louis (Klinger et al., 2016). When attempts at large-scale analyses have been conducted, those studies have been forced to draw upon more restricted data, typically those that consider only fatal encounters with police or offer conservative estimates of police homicides. Nationally, personnel with The Washington Post have assembled data from news outlets, law enforcement websites, social media, and independent police monitoring groups to create a database of fatal on-duty OIS incidents dating back to 2015 (Tate et al., 2016). Data were also collected identifying decedents’ race, mental health status, and whether they were armed during those OIS incidents (Tate et al., 2016). Exploring The Post data, research has assessed racial disparities among those fatally shot, specifically focusing on whether minorities were unarmed or not attacking police or others (Nix et al., 2017). In addition, Supplementary Homicide Report (SHR) data have been analyzed to assess the relationship of deadly force and structural elements of policing (Jacobs & O’Brien, 1998; Smith, 2004; Tennenbaum, 1994; Willits & Nowacki, 2014) though results derived from SHR data should be interpreted cautiously as it may underreport fatal incidents (Klinger, 2012; Klinger et al., 2016; Sherman & Langworthy, 1979).

Early research suggested that police use of deadly force varied greatly across jurisdictions and that Blacks were exceedingly involved in these incidents when compared with their percentage in the overall population (Fyfe, 1982). Several researchers have also argued that police officers are racially biased, either explicitly or implicitly (Hernández-Murillo & Knowles, 2004; Spencer et al., 2016), and view people of color, specifically Black or Hispanic males, as more dangerous than Whites (males or females). Spencer and colleagues (2016) suggested that the implicit biases held by police against Blacks lead to more frequent interventions, which may negatively influence the organizational legitimacy of police. In addition, if individuals believed that they had been targets of racial profiling, their attitudes toward the police were more negative (Weitzer & Tuch, 2002). Zhao et al. (2015) reached similar conclusions regarding citizen perceptions of racial bias in traffic stops utilizing citizen survey data from residents of Houston, TX. Research exploring deadly force and explicit or implicit racial bias, however, cannot be expanded nationally due to the dearth of available data related to OIS incidents to include demographics for both officers and suspects.

In the absence of these data, researchers have explored other means of assessing deadly force employed against the public. The use of crowdsourced data, data collected by media outlets (Campbell et al., 2018; Nix et al., 2017), and laboratory experiments (James et al., 2013, 2014) have become more common. These lines of research have reached starkly differing conclusions, however. Although some researchers assessing the real-world scenarios of OIS incidents have identified implicit racial bias on the part of police (Nix et al., 2017), others employing the same data have reached differing conclusions (Klinger & Slocum, 2017). Employing laboratory experiments simulating shoot/don’t shoot situations with civilians, James et al. (2014) found that even though participants held subconscious biases against Blacks, this subconscious bias did not translate to their behavior in simulations. Similar results were identified when conducting experiments with police, civilian, and military participants (James et al., 2013). The authors offered a number of possible explanations for the observed counter-bias, but additional research remains necessary to disentangle their findings. Public availability of data will be necessary to address the intertwined issues of OIS, racial bias, and police legitimacy.

Weitzer (2015) has noted that although there is considerable public support for police reform, successive scandals in policing have been attributed to “the police” (p. 479) rather than individual agencies, negatively impacting public perceptions. For example, successive deadly force incidents can erode public trust and negatively impact police legitimacy when complaints against officers are addressed internally, or when police officials defensively respond to the incidents (Fyfe, 1981; Kane, 2007; W. Katz, 2015; Terrill, 2016). Suchman (1995) noted that legitimacy is the cumulative perception that an organization has done what is expected and right. Organizations communicate adherence to these expectations by maintaining the symbolism and structure of other legitimate organizations within their field (DiMaggio & Powell, 1983; Donaldson, 1995; Meyer & Rowan, 1977; Meyer & Scott, 1983).

Crank and Langworthy (1992) mapped institutional theory onto U.S. police organizations by noting the symbolic nature of much of policing and its emphasis on attaining and maintaining legitimacy. Research has explored the power of legitimacy to police organizations, finding that a police chief would accept and respond to stakeholders’ perceptions of a problem when objectively one did not exist (C. Katz, 2001). By addressing community concerns, police can demonstrate that they are responsive to the needs of their environment (Crank, 2003). The maintenance of legitimacy is important, as is its connection with the environment and organizational disbanding (Maguire & King, 2007). Organizational disbanding, or organizational death, can occur for various reasons, but it is when an organization ceases to exist or is absorbed by another organization (Maguire & King, 2007). King (2014) found that organizations that do not heed the concerns of their environment are likely to experience legitimacy crises, which lead to disbanding, especially for smaller agencies. The current legitimacy crisis impacting policing may be addressed through increased transparency (Chanin & Espinosa, 2016). Establishing an accessible national deadly force reporting system would create greater openness to stakeholders suggesting increased levels of transparency in turn promoting police legitimacy (Fyfe, 1981; Kane, 2007; Koper, 2016).

This study evaluates overall levels of law enforcement transparency by exploring several key issues. First, the number of agencies making OIS data readily available to the public is identified. Second, how those data are made public is categorized. Finally, the types of variables provided are evaluated utilizing the framework of Klinger and colleagues (2016).

Data and Methods

There have been renewed calls for a national database to house information related to police use of deadly force and the solidification of ideal lists of OIS variables (FBI, 2019; Klinger et al., 2016), which suggests that the time is right to capture a baseline assessment of OIS data transparency across the United States. This research sought to identify agencies that have heeded these calls for transparency and to assess existing OIS data by exploring its classification and utility to researchers and policymakers. We restricted our search to municipal agencies and sheriffs’ offices servicing jurisdictions with a population of 100,000 or more based on the 2010 U.S. Census, which created an initial sample of 282 cities and 581 counties. Large jurisdictions were selected due to their prominence in U.S. policing research (Jacobs & O’Brien, 1998; Smith, 2004) and to the belief that these agencies are more likely to have the resources, both financial and personnel, necessary to maintain such a data collection and distribution effort. A mixed-methods approach was utilized to identify and quantify OIS data availability and classification. Organizations providing a form of OIS data were then compared on several structural variables drawn from the 2008 Census of State and Local Law Enforcement Agencies (CSLLEA). Below, we describe how we identified instances of data transparency and comparisons across data categories.

Our initial exploration of these 282 police agencies’, 581 sheriffs’ offices’, and municipal/county governments’ websites sought any data provided related to officer/deputy-involved shooting incidents. During the search, four police agencies were identified that did not provide law enforcement services, and one agency and its municipality did not have websites. The final sample of municipal police was made up of 277 agencies; this includes nine merged (municipal and county) agencies. A total of 33 sheriffs’ offices were eliminated from the search for the following reasons: provide civil enforcement only (16), corrections only (1), merged with a municipal agency (9), and no website (7). The final sample of sheriffs’ offices included 548 organizations.

Searches of municipal agencies’, sheriffs’ offices’, and municipal/county governments’ websites occurred between November 1, 2018, and January 31, 2019. Websites were visually searched for any references to OIS reporting or data. When search functions were available on individual agency and government websites, keyword searches were also conducted utilizing the terms: officer involved shooting, officer-involved shooting, OIS, deputy involved shooting, deputy-involved shooting, DIS, annual report, use of force, and open data. “Use of force” was included in our searches to identify those agencies that maintain an overall use of force database that identified officer/deputy-involved shootings as a category.

Publicly available OIS data took five primary forms: annual reports, OIS summary reports, individualized reports, press releases, and spreadsheet reports. These data can also be broadly categorized into three areas of information: incident, officer, and suspect. Of the 277 municipal agencies and 548 sheriffs’ offices in this sample, some form of data was available related to OIS incidents for 155 (56.0%) municipal agencies and 96 (17.5%) sheriffs’ offices. Table 1 summarizes the organizations and classification of OIS reporting. Annual and summary reports provided only the number of OIS incidents that occurred for an agency during a set period, typically by year. Summary reports were specific counts of OIS incidents that occurred within an agency; these reports did not contain information related to other activities carried out by the agency. Individual reports described a single OIS incident in varying levels of detail. It is important to note that half of the individualized reports are from the state of Texas (see Supplemental Appendix A) and contained 13 questions related to the incident from which limited data could be extracted (Texas Attorney General, 2016). Press releases varied in content provided, but these were frequently brief narratives that identified pertinent information related to the incident, officer(s), and suspect(s). Finally, spreadsheet reports were data provided by agencies at the incident level. These reports frequently presented data in the row (incident) and column (characteristics/variables) format. Locating data on an agency or government website does not immediately address the quality of those data provided. There is great variability in the data presented by each agency, years of availability, and accessibility of format.

Table 1.

Reporting of Municipal (n = 277) and Sheriffs’ (n = 548) OIS Incidents by Type.

Spreadsheet Report		Annual Report
Municipal (n = 32)	Sheriffs’ (n = 1)	Municipal (n = 31)	Sheriffs’ (n = 0)
Atlanta, GA	Lansing, MI	Abilene, TX	Newport News, VA
Austin, TX	Lincoln, NE	Billings, MT	Norfolk, VA
Baltimore, MD	Los Angeles Co., CA	Boise, ID	Peoria, AZ
Charleston, SC	Louisville, KY	Cape Coral, FL	Pittsburgh, PA
Charlotte, NC	Minneapolis, MN	Cleveland, OH	Reno, NV
Cincinnati, OH	New Orleans, LA	Elgin, IL	San Diego, CA
Dallas, TX	Norman, OK	Eugene, OR	San Jose, CA
Denver, CO	Orlando, FL	Gainesville, FL	Spokane, WA
Fayetteville, NC	Philadelphia, PA	Gilbert, AZ	St. Petersburg, FL
Fresno, CA	Richmond, VA	Greensboro, NC	Syracuse, NY
Hampton, VA	Santa Rosa, CA	Huntsville, AL	Toledo, OH
Hartford, CT	Scottsdale, AZ	Kansas City, KS	Topeka, KS
Henderson, NV	Seattle, WA	Lafayette, LA	Tulsa, OK
Houston, TX	South Bend, IN	Lexington, KY	Washington, D.C.
Indianapolis, IN	Tacoma, WA	Milwaukee, WI	Wilmington, NC
Jacksonville, FL	Tucson, AZ	Mobile, AL
Knoxville, TN
Individualized Report		Summary Report
Municipal (n = 15)	Sheriffs’ (n = 12)	Municipal (n = 12)	Sheriffs’ (n = 2)
Amarillo, TX	Midland Co., TX	Anaheim, CA	Newark, NJ
Beaumont, TX	Montgomery Co., TX	Arlington, TX	Oakland, CA
Comal Co., TX	Napa Co., CA	Chula Vista, CA	Phoenix, AZ
Deschutes Co., OR	Orange Co., NY	Clearwater, FL	Salt Lake City, UT
El Paso, TX	Portland, OR	Columbia, SC	San Diego Co., CA
Fort Worth, TX	Ramsey Co., MN	Fairfax Co., VA	San Francisco, CA
Irving, TX	Sacramento, CA	New York, NY	Springfield, MO
Killeen, TX	Salinas, CA
Lake Co., IL	San Antonio, TX
Linn Co., IA	Spokane Co., WA
Los Angeles, CA	Tarrant Co., TX
Lubbock, TX	Waco, TX
Madison, WI	Washington Co., OR
Mesquite, TX

Note. A list of agencies (n = 146, municipal n = 65 and sheriffs’ n = 81) providing press releases can be obtained by contacting the corresponding author. OIS = officer-involved shootings.

There are several reasons that the framework of Klinger and colleagues (2016) is used in this assessment rather than a newly constructed ideal list of variables. Although prior researchers have called for additional data collection related to OIS incidents, Klinger et al. (2016) specifically identified those variables that should be included in future data collection efforts. They have called for the reporting of sex, age, race, ethnicity, injuries, and sources of injuries for all participants in shooting incidents. By all participants, this includes all suspects, all officers (not only those that discharged a firearm), and all other persons involved to include hostages, victims, and bystanders (Klinger et al., 2016). In addition to demographic information for all participants, Klinger et al. (2016) proposed specific information be collected regarding suspects and officers. Suspect information should also include criminal history, mental health status, use of intoxicants, and type of weapon possessed. Officer information should also include rank, years in service, number of prior shootings, assignment at time of shooting, type of weapon fired, caliber of weapon fired, number of rounds fired, and less lethal weapons deployed prior to use of deadly force. Finally, the location and participating agencies should be identified. The recent data collection efforts launched by the FBI (2019) largely replicate those variables identified by Klinger and colleagues. This suggests an accepted baseline by those in these types of law enforcement agencies and the beginning of the institutionalization of this data collection process. The framework employed attempts to balance researchers’ demands for useful data and officers’ demands of efficiency in the field.

Data from the 2008 CSLLEA were utilized to compare agencies that provided information regarding OIS incidents. Table 2 includes demographic information for all reporting agencies and by OIS classification. Due to the vast range for agency size, specifically the difference between the largest and next agency in each category, and the resulting inflation of the mean, Table 2 reports two measures of agency size. The first measure identified by a superscript “a” included the largest organization from each category. Agencies removed in the second reporting of agency size included the Los Angeles County Sheriff’s Department (spreadsheet report), Washington DC Metropolitan Police Department (annual report), Los Angeles Police Department (LAPD; individual report), New York Police Department (summary report), and Chicago Police Department (press release). Utilizing one-way analysis of variance (ANOVA), organizations were compared by OIS classification on four characteristics available in the 2008 CSLLEA: agency size, civilianization, task scope, and interagency collaborations. Agency size was defined as the number of full-time sworn and nonsworn employees employed by the organization. Larger organizations are expected to have greater resources and would therefore be more likely to provide information related to OIS incidents. The proportion of nonsworn employees is referred to as civilianization, which is a measure of employee specialization. Providing OIS data to the public would require knowledge and technical skills beyond that of standard law enforcement duties. Task scope was a cumulative index that ranged from 0 to 39 and measured the number of functions that an organization has primary responsibility for in their jurisdiction (e.g., traffic enforcement, homicide investigations, routine patrol, operation of a jail). The influence of task scope is similar to that of agency size; organizations responsible for a greater number of functions are anticipated to have access to greater resources allowing for the provision of more information about OIS incidents. Finally, interagency collaboration was also a cumulative index ranging from 0 to 6. It measured the number of task forces in which organizations participated. It is believed that organizations with greater levels of collaboration prioritize the maintenance of relationships, which is likely accomplished through communication and mutual transparency. In addition, 2008 CSLLEA data were employed to identify regional distribution of reporting agencies based on U.S. Census definitions (see Table 3).

Table 2.

Organizational Characteristics of OIS Reporting Agencies.

Characteristics	N	Minimum	Maximum	M	Median	SD
All reporting agencies
Agency size	251	64.0	49,926.0	1,247.85	492.0	3,608.53
Civilianization	251	0.000	0.737	0.323	0.287	0.161
Task scope (0–39)	249	10.0	37.0	25.35	25.0	4.82
Interagency collaborations (0–6)	249	0.0	6.0	3.37	3.0	1.42
Spreadsheet report agencies
Agency size^a	33	183.0	16,602.0	2,083.45	1,266.0	3,116.65
Agency size	32	183.0	7,463.0	1,629.75	1,195.5	1,736.30
Civilianization	32	0.050	0.440	0.232	0.231	0.100
Task scope (0–39)	31	19.0	33.0	23.87	23.0	3.46
Interagency collaborations (0–6)	31	1.0	6.0	3.87	4.0	1.23
Annual report agencies
Agency size^a	31	164.0	4,647.0	859.26	505.0	962.33
Agency size	30	164.0	2,739.0	733.00	495.5	668.43
Civilianization	30	0.075	0.409	0.222	0.202	0.077
Task scope (0–39)	30	15.0	32.0	22.77	23.0	3.62
Interagency collaborations (0–6)	30	0.0	6.0	3.70	4.0	1.49
Individual report agencies
Agency size^a	27	108.0	12,999.0	1,099.00	505.0	2,449.66
Agency size	26	108.0	2,564.0	641.31	487.5	598.83
Civilianization	26	0.112	0.724	0.353	0.274	0.175
Task scope (0–39)	26	20.0	33.0	26.27	26.0	3.68
Interagency collaborations (0–6)	26	0.0	6.0	3.08	3.0	1.60
Summary report agencies
Agency size^a	14	338.0	49,926.0	4,824.86	696.5	13,049.13
Agency size	13	338.0	4,432.0	1,355.54	607.0	1,386.09
Civilianization	13	0.095	0.665	0.275	0.269	0.145
Task scope (0–39)	13	19.0	34.0	25.08	24.0	4.41
Interagency collaborations (0–6)	13	1.0	6.0	3.85	4.0	1.35
Press release agencies
Agency size^a	146	64.0	14,320.0	826.02	367.0	1,470.81
Agency size	145	64.0	5,688.0	732.96	366.0	951.35
Civilianization	145	0.000	0.737	0.366	0.322	0.167
Task scope (0–39)	145	10.0	37.0	25.91	26.0	5.24
Interagency collaborations (0–6)	145	0.0	6.0	3.20	3.0	1.35

Note. OIS = officer-involved shootings.

Largest agency in reporting category included.

Table 3.

Regional Distribution of Agencies by OIS Data Type.

Region	Spreadsheet Report (n = 33)	Annual Report (n = 31)	Individual Report (n = 27)	Summary Report (n = 14)	Press Release (n = 146)
Northeast (n = 14)	2	2	1	2	7
Midwest (n = 49)	6	6	4	1	32
South (n = 96)	16	14	14	4	48
West (n = 92)	9	9	8	7	59

Note. OIS = officer-involved shootings.

Findings

In looking at all 277 municipal police agencies and 548 sheriffs’ offices in the U.S. serving municipalities and counties with 100,000 or more residents, slightly more than half (56.0%, n = 155) of municipal agencies and less than 20% (n = 96) of sheriffs’ offices provided data on the agency’s or municipal government’s website related to OIS incidents. When breaking down the five categories identified, 11.6% of municipal agencies and 0.18% of sheriffs’ offices publicly disclosed data related to their OIS incidents at the incident level (i.e., spreadsheet reporting). No sheriffs’ offices were identified that provided OIS data in their annual reports, whereas 11.2% of municipal agencies provided data in this format. Approximately, 4.3% of municipal agencies and 0.4% of sheriffs’ offices presented aggregated summary reports of OIS incidents. Individualized shooting reports were provided by about 5.4% of municipal agencies and 2.2% of sheriffs’ offices. Finally, 23.5% of municipal agencies and 14.8% of sheriffs’ offices communicated information related to their OIS incidents through press releases. Overall, this amounts to about 44% of municipal police agencies and 82% of sheriffs’ offices that service populations greater than 100,000 who do not make data publicly available related to OIS incidents through their website. Applying these findings to the 15,328 law enforcement agencies in the United States (Hyland, 2018), it suggests that most agencies are not providing the transparency necessary to inform the communities that they serve regarding uses of deadly force, specifically OIS.

The organizational characteristics presented in Table 2 show variation across the five reporting categories. ANOVA was performed utilizing Tukey’s test to assess organizational differences between reporting categories. Results showed a significant effect for agency size (F = 5.869, p < .001, partial η² = .089), civilianization (F = 9.735, p < .001, partial η² = .139), task scope (F = 3.811, p = .005, partial η² = .060), and interagency collaborations (F = 2.660, p = .033, partial η² = .043). Effect sizes for each organizational characteristic are small, however. Tukey’s post hoc test was utilized to identify which reporting categories were different. Agency size of spreadsheet reporting agencies was greater and differed significantly from annual report, individualized report, and press release agencies. Individualized report and press release agencies employ a greater proportion of civilians than do spreadsheet and annual reporting agencies. Individualized report and press release agencies have primary responsibility for a greater mean number of functions compared with annual report agencies. No individual reporting groups differed significantly on interagency collaborations. The regional breakdown of those organizations providing data about OIS incidents is as follows: Northeast (5.6%, n = 14); Midwest (19.5%, n = 49); South (38.2%, n = 96); and West (36.7%, n = 92). Within these regions, Table 3 identifies variation in the availability of OIS information. The findings related to information classification are discussed in the context of this sample of 155 municipal agencies and 96 sheriffs’ offices that provided information related to OIS. Each of these five types of reporting will be assessed through the Klinger et al. (2016) framework described previously.

Annual and Summary Reports

Although annual and summary reports do provide some detail about a jurisdiction’s prevalence of and response to OIS incidents, the way the data are presented does not provide opportunities for analyses (bivariate or multivariate). The annual report category often briefly identifies only the raw number on OIS incidents in a given year. No other mention of these incidents are included in the annual reports. Summary reports also often only identify the total number of shootings in the report. Although a useful tool for the department and illustrative of increased transparency, the lack of incident-level variables limits its applicability to research endeavors. These organizations, however, should be commended for a level of openness not seen in a large segment of departments across the nation. Neither group approaches the minimum level of detail for OIS data collection proposed by Klinger et al. (2016). Overall, researchers and the public require additional information beyond that provided in annual and summary reports to assess the state of OIS incidents in law enforcement organizations, as well as to formulate and evaluate public policy related to these incidents.

Individualized Reports

The individualized reporting agencies group identified in Table 1 is made up of 15 municipal agencies and 12 sheriffs’ offices. More than half of this group comprises Texas agencies; the Texas legislature passed a state law (Texas Code of Criminal Procedure Art. 2.139, 2015) in which certain information about OIS must be documented. As such, Texas has created a standardized form for agencies to submit to the Attorney General of Texas. While conducting the research, Texas was the only state identified attempting to standardize this process across law enforcement agencies using legislation. Similar to other individualized reporting agencies, the Texas form includes several variables identified by Klinger and colleagues (2016) to include as follows: agency, location, demographics of officer and suspect, injury to suspect, and whether the suspect had a deadly weapon. Overall, these individualized shooting reports lack a number of the suspect- and offender-specific data points identified in the framework. These reports can be of limited use to researchers, but they lack the data necessary to broadly inform public policy.

Two agencies in the individualized report category that merit their own descriptions due to the rich qualitative data that their files provide are the LAPD and Portland Police Department (PD; Oregon). The LAPD documents the investigation of each OIS and opens the records of the Los Angeles Board of Police Commissioners for public consumption. In addition, when Portland PD conducted death investigations, documentation to include grand jury transcripts is provided in the PDF format, which was unique to Portland PD in this sample of agencies. It should be noted, however, that this level of documentation and transparency was applicable only to death investigations and not all OIS. Although these two agencies provide extensive data that align with the Klinger et al. (2016) framework, the limitation of these types of reports is the immense commitment of time and resources to their coding before these cases could be incorporated into quantitative analyses. In addition, the reports do not follow a standardized format, meaning that some reports may meet the minimums identified by Klinger and colleagues, whereas others do not.

Press Releases

The largest reporting category for both municipal agencies (n = 65, 23.47%) and sheriffs’ offices (n = 81, 14.78%) was press releases. These press releases provided brief narratives related to OIS incidents. Frequently, press releases contained the date, time, and location of the incident in addition to several sentences that described what led to the shooting. Press releases varied on the amount of information provided related to officers and suspects involved in the shooting incidents. Variation in press release information was within and across organizations, and no law enforcement organizations provided all information identified in the framework of Klinger et al. (2016).

Spreadsheet Reporting

Thirty-two municipal police agencies and one sheriff’s office make up the spreadsheet reporting category. All these agencies were included in this category because they provided data, primarily in spreadsheets available for download or in a column and row form that can be easily transferred to a spreadsheet-type application. Although organizations in the spreadsheet reporting category frequently provided more data in line with the minimum variables identified by Klinger et al. (2016), this was not an indicator of the quality of those data for the purposes of analyses. Although many of the departments utilize Microsoft Excel to present the OIS data in downloadable form, several provided their data in the PDF format or in a web-based spreadsheet-like format. In addition, Austin PD (Texas) provided datasets broken down, not by year, but by suspect, officer, and incident.

Several municipal police agencies provided data that were robust and detailed enough for quantitative researcher. Austin PD provided the most variables (19) identified by the Klinger et al. framework, whereas Houston PD (Texas) had the greatest overall number of OIS incidents available for analyses across 13 framework variables. Other agencies provide limited information related to OIS incidents such as the location and agency responding. The average number of variables identified across the 33 organizations was 9.7.

A summary of variables provided by agencies within the framework of Klinger and colleagues (2016) is presented in Table 4 demonstrating the broad variation in alignment between the framework and data collected by the 33 (32 municipal and one sheriff) agencies identified. Agency involved was the only universally provided variable from the framework, and none of the 33 agencies provide data for all variables suggested by Klinger and colleagues. The most frequently provided information in the spreadsheet report category identified the location of the incident (address or geocoding), officer race and sex, as well as suspect race, sex, and injuries. No agencies, however, provided information related to suspects’ criminal history, and only one agency identified whether an officer was previously involved in an OIS, caliber of weapons fired, whether less than lethal force was deployed prior to the shooting, source of suspect’s injury, and suspect’s mental health. In addition, no agencies provided information identifying all officers, suspects, or witnesses to the incidents. Most frequently, agencies provided data related to location and suspect demographics; fewer agencies provided the same level of data related to officers involved in the incidents.

Table 4.

OIS Data in the Framework of Klinger and Colleagues (2016).

Variable	Number of Agencies	Percentage of Agencies
Agency	33	100.00
Location—address	26	78.79
Location—geocoding	13	39.39
Suspect
Sex	25	75.76
Age	16	48.48
Race	25	75.76
Ethnicity	16	48.48
Injuries	23	69.70
Source of injuries	1	3.03
Criminal history	0	0.00
Mental health status	1	3.03
Intoxicant status	2	6.06
Weapon	13	39.39
Officer
Sex	22	66.67
Age	14	42.42
Race	22	66.67
Ethnicity	14	42.42
Injuries	14	42.42
Rank	3	9.09
Tenure	13	39.39
Prior OIS incident	1	3.03
Assignment	14	42.42
Weapon	6	18.18
Weapon caliber	1	3.03
Number of rounds fired	4	12.12
Less lethal force deployed	1	3.03

Note. OIS = officer-involved shootings.

Table 4 demonstrates the lack of consistency with which agencies provide data aligned with the Klinger et al. (2016) framework. To provide meaningful and objective information to policymakers, researchers require access to consistent data across many agencies. Although the FBI has begun a data collection effort across the United States that is largely based on the framework discussed above, it has not been institutionalized throughout even the limited number of police organizations providing data regarding OIS incidents.

Discussion

The current assessment of police agency, municipal/county government, and sheriffs’ offices’ websites sought to explore the transparency of law enforcement organizations serving jurisdictions with a population of 100,000 or more by assessing the type and classification of data shared related to OIS incidents throughout the United States. Based on the findings of this research, it can be stated that the state of publicly available OIS information is limited with a small proportion of law enforcement organizations providing any data related to OIS incidents. Importantly, data presented by law enforcement agencies lack continuity as well. There is great variation in the type, classification, and quality of data presented that limits its utility for the purposes of research as well as policy creation and evaluation. The lack of transparency and limited functionality of available data no doubt contributes to the current challenges to police legitimacy in the United States (Chanin & Espinosa, 2016), and it creates opportunity for other groups, such as The Washington Post, The Guardian, Mapping Police Violence, Killed by Police, and Fatal Encounters, to create their own databases and tracking efforts related to OIS incidents.

Klinger and colleagues’ (2016) five broad areas and the variables included for data collection related to OIS incidents are the culmination of ideal lists created and refined by researchers over time (Fyfe, 1978; Kane, 2007; Reiss, 1971), and they have been implemented into practice by the FBI (2019). The current state of OIS data in the United States is highly disorganized. Although we have identified a number of law enforcement organizations that provide publicly available data related to their OIS incidents (see Table 1), a much smaller portion provided data that could easily be used for analyses. A smaller portion still provides data of sufficient quality and quantity to be useful to researchers and policymakers.

As Weitzer (2015) has noted, the response to OIS incidents has largely focused on issues of race between officers and suspects involved. This has led to a highly politicized environment that suggests racial bias on the part of law enforcement and racial disparity in the outcomes of interactions with the communities served. The current state of publicly available OIS data does not allow for adequate exploration of the issues of racial bias and disparity. For example, those organizations that identify the number of OIS incidents experienced in their annual or summary reports and make those reports publicly available achieve a rare level of transparency in the field. The identification of a raw number of OIS incidents, however, does nothing to inform the law enforcement organization or community about the lessons that could have been learned from those shootings. Annual and summary reports provide no data that would contribute to research or improve policy.

The data provided in individualized shooting reports and press releases suffer from similar limitations. As noted above, LAPD and Portland PD provided extensive investigative files that could be utilized by researchers. From these files, variables could be extracted and coded to create quantitative data. This would be a time-consuming process, and it is not clear how data contained within these files aligned with the framework of Klinger and colleagues (2016). The investigative files that were presented as individualized shooting reports, however, contained a considerable amount of information that would likely be useful to researchers and policymakers. Content analysis could also be completed with press releases, though the return on investment for press releases is likely small. Data contained in press releases appeared to be included at the discretion of the individual writer rather than organizational policy.

Although the individual shooting reports from Texas law enforcement organizations did not provide sufficient data in the framework of Klinger and colleagues (2016) to be of use to researchers or policymakers, these reports were the most consistent identified during the research. As of 2015, the Texas Code of Criminal Procedure Article 2.139 has required a report be filed with the Texas Attorney General in the event of an officer-involved injury or death. Texas appeared to be the only state that required a standardized format in reporting OIS incidents as there was considerable variation in the way OIS information was presented within other states. The Texas law was a positive step toward transparency and standardized OIS reporting, but they suffer from not being able to capture all intentional discharges of police firearms, as only discharges resulting in injury or death must be reported. This limits the value of these reports for the purpose of training, research, and policy as well.

Those law enforcement organizations in the spreadsheet report category also provided data that were inconsistent and frequently of poor quality. With the exception of several organizations, these data would not be useful to researchers or policymakers. No organizations provided all of the “minimum” variables identified by the Klinger et al. (2016) framework. Although these organizations frequently provided demographic data related to suspects, factors that may have contributed to the incidents (e.g., mental health status, intoxicant status, criminal history) were rarely captured. Including these variables in data collection would at a minimum allow organizations to conduct a more thorough review of incidents to inform their own policies and training. Fewer organizations provided the same level of demographic data about their officers. Although this may be due to collective bargaining agreements or other policies, it limits the research that can be conducted with the available data.

Attempting to apply the framework of Klinger et al. (2016) to the available data demonstrated a clear lack of consistency, but it reinforced the need for standardization of reporting and data collection among agencies. The uniform reporting of Texas law enforcement organizations suggests that this standardization is possible. Standardized reporting would not only benefit transparency, but it would also allow organizations to evaluate their own training, policies, and other relevant information in relation to other law enforcement organizations (Klinger et al., 2016). A national guide is necessary to direct a large-scale effort to capture useful data across the entire U.S. policing system. That initial step forward appears to have been taken with the launch of the FBI National Use-of-Force Data Collection (January 1, 2019), which operates in conjunction with the Uniform Crime Reporting (UCR) program (FBI, 2019). There are several benefits to the FBI’s efforts. First, data are collected when an individual is killed, seriously injured, or when individuals are shot at and missed. Second, those data being collected largely align with the framework proposed by Klinger and colleagues (2016), suggesting a link between academic and practitioner understanding of OIS incidents. Third, the FBI is viewed as a premier law enforcement organization in the United States. The FBI possesses the legitimacy necessary to influence the entire institutional field of policing in the United States. Phillips and Jiao (2016) have explained the isomorphic process associated with LAPD’s development of auditing knowledge in response to a Department of Justice consent decree. At the time of the consent decree, LAPD personnel found no other models within the institutional environment to base their own auditing capabilities. After developing auditing knowledge and capabilities, LAPD became the model agencies for other law enforcement organizations (Phillips & Jiao, 2016). We believe that the collection of OIS data will be subject to a similar isomorphic process and eventual institutionalization as a result of the efforts of the FBI. We hope to see the FBI make these data publicly available to researchers and believe that their use will become commonplace like that of the UCR. In addition, broad reporting (i.e., transparency) of these data could satisfy external groups that have been critical of police responses to OIS incidents.

There were a number of limitations associated with this research. The first is that our assessment of OIS reporting was restricted to agencies serving populations of 100,000 or more (277 cities and 548 counties). This sample of 825 agencies represents approximately 5.4% of the 15,328 agencies identified by the Bureau of Justice Statistics (Hyland, 2018). In addition, we did not assess the type of government structure (individual executive vs. a board/council) of the municipalities included in our assessment. It should be noted, however, that the elected nature of sheriffs’ positions did not appear to incentivize transparency. Jurisdictions servicing smaller populations, state agencies, special jurisdictions, federal agencies, and constables/marshals were not included in this research. In addition, the terms utilized in keyword searches may have inadvertently restricted our findings. A search of broader keywords (e.g., deadly force or shooting) may have returned added results not identified by targeted search terms employed in this research. A second limitation of this research is related to the differences between the reporting categories identified through the one-way ANOVA with Tukey’s test. Due to the small sample sizes and extreme ranges for agency size, civilianization, and task scope, it is not clear whether these significant differences reflect true patterns or spurious differences. A third major limitation of our analyses is the fluidity of information posted on the internet and specifically agencies’ websites. As leadership within these agencies changes, their reporting transparency practices may also change. Therefore, it is possible that after January 31, 2019 agencies included in this assessment could restrict access to their data. This situation has occurred with the individual shooting reports collected by the Texas Attorney General’s office. Previously, these reports were accessible to any person viewing the Attorney General’s website; as of January 31, 2019, these reports require a user name and password to access. This fluidity cuts both ways, however, as additional agencies may decide to provide OIS data to the public; the recent national collection effort by the FBI may encourage disclosure throughout the field of law enforcement. Fourth, our analyses and discussion have focused on the minimum framework identified by Klinger and colleagues (2016). There was a broad spectrum of variables identified and collected by organizations in the spreadsheet reporting category (e.g., weather and lighting conditions, duty status [on/off], whether English was a suspect’s second language) that we have been unable to communicate here but would be of interest to law enforcement organizations. Finally, this study did not account for agencies without OIS incidents; not all agencies will experience an OIS incident. Therefore, a portion of those agencies classified as “none” may have had no relevant information to provide. If this was the case for an agency, it may be information that they wish to highlight on their website.

Conclusion

This research found that many municipal police agencies and sheriffs’ offices serving large populations are not transparent in regards to OIS incidents. This lack of access to data limits the ability of researchers and policymakers to make informed decisions about public policy. The ability to assess the effectiveness of policy and hold individuals and organizations accountable is also restricted. Our findings demonstrated that, even when law enforcement organizations provide data related to OIS incidents, it is of limited use for research and policy, due to its frequent inconsistencies. These inconsistencies do not allow researchers or policymakers to objectively assess how OIS impact differing groups across society. Future research should explore the influence that OIS transparency has on the perceptions of community stakeholders. Is access to those data a key factor in improving law enforcement legitimacy in the United States? In addition to the minimum data collection efforts identified by Klinger et al. (2016), researchers should explore the diffusion of the FBI’s innovative national collection efforts throughout the field of law enforcement. Doing so will allow researchers to explore the viability and benefit of collecting, sharing, and analyzing data across agencies. This research should lead to better public policy and allow for greater understanding of policing and its legitimacy in the United States.

Supplemental Material

appendix_a – Supplemental material for An Assessment of Officer-Involved Shooting Data Transparency in the United States

Supplemental material, appendix_a for An Assessment of Officer-Involved Shooting Data Transparency in the United States by Matthew C. Matusiak, Michael R. Cavanaugh and Matthew Stephenson in Journal of Interpersonal Violence

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Matthew C. Matusiak

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Matthew C. Matusiak is an associate professor in the Department of Criminal Justice at the University of Central Florida. He earned his PhD in criminal justice from Sam Houston State University in 2013. His primary areas of research interest include police organizational structure, executive police leadership, and organizational theory.

Michael R. Cavanaugh is an associate professor in the Department of Criminal Justice at the University of Houston–Downtown. He received his PhD in criminal justice from Sam Houston State University in 2012. His research interests include officer-involved shootings, campus carry, and a wide array of legal issues.

Matthew Stephenson is a doctoral student in the Department of Criminal Justice at the University of Central Florida. He received his MA in criminal justice from Temple University in 2014. His research interests include the use of new and innovative technology in policing.

References

Alpert

G. P.

(2016). Toward a national database of officer-involved shootings: A long and winding road. Criminology & Public Policy, 15(1), 237–242.

Campbell

B. A.

Nix

Maguire

E. R.

(2018). Is the number of citizens fatally shot by police increasing in the post-Ferguson era? Crime & Delinquency, 64(3), 398–420.

Chanin

Espinosa

(2016). Examining the determinants of police department transparency: The view of police executives. Criminal Justice Policy Review, 27(5), 498–519.

Crank

J. P.

(2003). Institutional theory of police: A review of the state of the art. Policing: An International Journal of Police Strategies & Management, 26(2), 186–207.

Crank

J. P.

Langworthy

(1992). An institutional perspective of policing. The Journal of Criminal Law & Criminology, 83(2), 338–363.

DiMaggio

P. J.

Powell

W. W.

(1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160.

Donaldson

(1995). American anti-management theories of organizations: A critique of paradigm proliferation. Cambridge University Press.

Federal Bureau of Investigation. (2019). National use-of-force data collection. https://www.fbi.gov/services/cjis/ucr/use-of-force

Fyfe

J. J.

(1978). Shots fired: An examination of New York City Police firearms discharges. University Microfilms International.

10.

Fyfe

J. J.

(1980). Geographic correlates of police shooting: A microanalysis. Journal of Research in Crime & Delinquency, 17(1), 101–113.

11.

Fyfe

J. J.

(1981). Observations on police deadly force. Crime & Delinquency, 27(3), 376–389.

12.

Fyfe

J. J.

(1982). Blind justice: Police shootings in Memphis. Journal of Criminal Law & Criminology, 73(2), 707–722.

13.

Fyfe

J. J.

(1988). Police use of deadly force: Research and reform. Justice Quarterly, 5(2), 165–205.

14.

Fyfe

J. J.

(2002). Too many missing cases: Holes in our knowledge about police use of force. Justice Research & Policy, 4(1–2), 87–102.

15.

Hernández-Murillo

Knowles

(2004). Racial profiling or racist policing? Bounds tests in aggregate data. International Economic Review, 45(3), 959–989.

16.

Hyland

(2018). Full-time employees in law enforcement agencies, 1997–2016 (NCJ 251762). Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

17.

Jacobs

O’Brien

R. M.

(1998). The determinants of deadly force: A structural analysis of police violence. American Journal of Sociology, 103(4), 837–862.

18.

James

Klinger

Vila

(2014). Racial and ethnic bias in decisions to shoot seen through a stronger lens: Experimental results from high-fidelity laboratory simulations. Journal of Experimental Criminology, 10(3), 323–340.

19.

James

Vila

Daratha

(2013). Results from experimental trials testing participant responses to White, Hispanic and Black suspects in high-fidelity deadly force judgement and decision-making simulations. Journal of Experimental Criminology, 9(2), 189–212.

20.

Kane

R. J.

(2007). Collect and release data on coercive police actions. Criminology & Public Policy, 6(4), 773–780.

21.

Katz

C. M.

(2001). The establishment of a police gang unit: An examination of organizational factors. Criminology, 39(1), 37–74.

22.

Katz

(2015). Enhancing accountability and trust with independent investigations of police lethal force. Harvard Law Review, 128(6), 235–245.

23.

King

W. R.

(2014). Organizational failure and the disbanding of local police agencies. Crime & Delinquency, 60(5), 667–692.

24.

Klinger

D. A.

(2012). On the problems and promise of research on lethal police violence: A research note. Homicide Studies, 16(1), 78–96.

25.

Klinger

Rosenfeld

Isom

Deckard

(2016). Race, crime, and the micro-ecology of deadly force. Criminology & Public Policy, 15(1), 193–222.

26.

Klinger

D. A.

Slocum

L. A.

(2017). Critical assessment of an analysis of a journalistic compendium of citizens killed by police gunfire. Criminology & Public Policy, 16(1), 349–362.

27.

Koper

C. S.

(2016). Advancing research and accountability on police use of deadly force. Criminology & Public Policy, 15(1), 187–191.

28.

Maguire

E. R.

King

W. R.

(2007). The changing landscape of American police organizations. In Schafer

J. A.

(Ed.), Policing 2020: Exploring the future of crime, communities, and policing (pp. 337–371). Federal Bureau of Investigation.

29.

Meyer

J. W.

Rowan

(1977). Institutional organizations: Formal structure as myth and ceremony. The American Journal of Sociology, 83(2), 340–363.

30.

Meyer

J. W.

Scott

W. R.

(1983). Centralization and the legitimacy problems of local government. In Meyer

J. W.

Scott

W. R.

(Eds.), Organizational environments: Ritual and rationality (pp. 199–215). SAGE.

31.

Nix

Campbell

B. A.

Byers

E. H.

Alpert

G. P.

(2017). A bird’s eye view of civilians killed by police in 2015: Further evidence of implicit bias. Criminology & Public Policy, 16(1), 309–340.

32.

Phillips

J. R.

Jiao

A. Y.

(2016). Institutional isomorphism and the federal consent decree: The case of the Los Angeles police department. Policing: An International Journal of Police Strategies & Management, 39(4), 756–772.

33.

Reiss

(1971). Police and the public. Yale University Press.

34.

Robin

G. D.

(1963). Justifiable homicide by police officers. Journal of Criminal Law, Criminology, & Police Science, 54(2), 225–231.

35.

Sherman

L. W.

Langworthy

R. H.

(1979). Measuring homicide by police officers. Journal of Criminal Law & Criminology, 70(4), 546–560.

36.

Smith

B. W.

(2004). Structural and organizational predictors of homicide by police. Policing: An International Journal of Police Strategies & Management, 27(4), 539–557.

37.

Sparger

J. R.

Giacopassi

D. J.

(1992). Memphis revisited: A reexamination of police shootings after the Garner decision. Justice Quarterly, 9(2), 211–225.

38.

Spencer

K. B.

Charbonneau

A. K.

Glaser

(2016). Implicit bias and policing. Social and Personality Psychology Compass, 10(1), 50–63.

39.

Suchman

M. C.

(1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610.

40.

Tate

Jenkin

Rich

Muyskens

Elliott

Mellnik

Williams

(2016, July 7). How The Washington Post is examining police shootings in the United States. The Washington Post. https://www.washingtonpost.com/national/how-the-washington-post-is-examining-police-shootings-in-the-united-states/2016/07/07/d9c52238-43ad-11e6-8856-f26de2537a9d_story.html

41.

Tennenbaum

A. N.

(1994). The influence of the Garner decision on police use of deadly force. Journal of Criminal Law & Criminology, 85(1), 241–260.

42.

Terrill

(2016). Deadly force: To shoot or not to shoot. Criminology & Public Policy, 15(2), 491–496.

43.

Texas Attorney General. (2016). Peace officer involved shootings, 2015 annual report. http://texasattorneygeneral.gov/cj/peace-officer-involved-shooting-report

44.

Texas Code of Criminal Procedure Art. 2.139. Reports required for officer-involved injuries or deaths (2015).

45.

Weitzer

(2015). American policing under fire: Misconduct and reform. Social Science & Public Policy, 52(5), 475–480.

46.

Weitzer

Tuch

S. A.

(2002). Perceptions of racial profiling: Race, class, and personal experience. Criminology, 40(2), 435–456.

47.

White

M. D.

(2001). Controlling police decisions to use deadly force: Reexamining the importance of administrative policy. Crime & Delinquency, 47(1), 131–151.

48.

White

M. D.

(2002). Identifying situational predictors of police shootings using multivariate analysis. Policing: An International Journal of Police Strategies & Management, 25(4), 726–751.

49.

White

M. D.

(2006). Hitting the target (or not): Comparing characteristics of fatal, injurious, and noninjurious police shootings. Police Quarterly, 9(3), 303–330.

50.

White

M. D.

(2016). Transactional encounters, crisis-driving reform, and the potential for a national police deadly force database. Criminology & Public Policy, 15(1), 223–235.

51.

Williams

H. E.

Bowman

S. W.

Jung

J. T.

(2019). The limitations of governmental databases for analyzing fatal officer-involved shootings in the United States. Criminal Justice Policy Review, 30(2), 201–222.

52.

Willits

D. W.

Nowacki

J. S.

(2014). Police organisation and deadly force: An examination of variation across large and small cities. Policing & Society, 24(1), 63–80.

53.

Zhao

Lai

Ren

Lawton

(2015). The impact of race/ethnicity and quality of life policing on public attitudes toward racially biased policing and traffic stops. Crime & Delinquency, 61(3), 350–374.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB