To examine differences in use of force by police patrol and specialized units, and the impact of body-worn cameras (BWCs) on use of force in these groups.
We use administrative data from the Tempe (AZ) Police Department collected during a randomized controlled trial (RCT) of BWCs. t tests of means and ARIMA models were constructed to analyze unit-level variation in use of force.
Tempe officers in specialized units use substantially more force than patrol officers. BWCs had no impact on use of force among patrol officers but were associated with a significant decline in force among specialty unit officers who received BWCs in the second phase of the study.
Unit-level variations in force can have implications for selection, training, and other areas of police practice. Additionally, our findings show the necessity of accounting for group variation within departments when assessing the impact of BWCs on outcomes like use of force.
There is a long history of racial injustice in American policing, with use of force at the core (White and Fradella 2016). In 1968, the National Advisory Commission on Civil Disorders concluded that “deep hostility between police and ghetto communities” was a primary cause of the riots that occurred during that decade (The National Advisory Commission on Civil Disorders 1968, p. 157). Race riots sparked by police use of force also occurred in 1980 (Miami), 1992 (Los Angeles), 1996 (St. Petersburg), and 2001 (Cincinnati).
A series of more recent incidents has again demonstrated the powerful consequences of police use of force against minority citizens. On August 9, 2014, officer Darren Wilson shot and killed Michael Brown in Ferguson, Missouri. Riots occurred the day of the shooting and again in November 2014 when the grand jury announced its decision to not indict Wilson. Five months later, rioting occurred in Baltimore following the death of Freddie Gray. Public protests and demands for police reform following police killings of Brown, Gray, and other minority citizens—including Eric Garner, Walter Scott, Samuel Dubose, and Tamir Rice—contributed to then President Obama creating the President’s Task Force on twenty-first Century Policing, the first presidential commission to examine police use of force in more than 40 years.
Given its devastating consequences, researchers have devoted significant attention to understanding police use of force decision-making, and over the last six decades, a robust body of literature on the causes and correlates of the phenomena has developed (Bittner 1970; Fyfe 1988; Shjarback and White 2016; Terrill and Mastrofski 2002). Scholars have identified three sets of variables—situational, social, and organizational factors—that significantly influence officer use of force (White and Klinger 2012). However, there are still notable gaps in our understanding of use of force (Klinger 2008; White 2016). In particular, research is limited as to whether force rates vary across officer working groups. Many departments have become highly specialized, with units assigned to handle specific problems or people, such as traffic, SWAT, and anti-crime. The majority of US police departments have specialized units (Reaves 2015), and given the distinctive missions of some specialized units, it is reasonable to hypothesize that officers in those assignments would be more likely to use force than officers assigned to patrol (Gaub et al. 2020). Only two studies have examined this question empirically, and the results are far from definitive (Brandl et al. 2001; Williams and Westall 2003). As a result, the degree to which the nature and prevalence of use of force may vary among patrol and specialized units remains an open question.
Relatedly, police body-worn cameras (BWCs) have emerged as a tool to potentially alleviate the crisis surrounding police use of force (White 2014). Nearly 20 studies have examined the impact of BWCs on use of force by police, and the findings are mixed (White et al. 2019). Some have documented large declines in use of force following BWC deployment (Ariel et al. 2015; Jennings et al. 2015), while others have reported no impact (Yokum et al. 2017). Much like the larger literature on use of force, BWC researchers have failed to examine force rates at the unit level. Rather, studies have calculated broad trends in use of force across departments. In effect, the mixed findings on BWC impact may, at least partially, be explained by the differential response to BWCs among officers in specialty and patrol units. The current study takes a step toward addressing these gaps in both areas of research by (1) looking more deeply at force rates across specialty and patrol units in the Tempe (AZ) Police Department; and (2) assessing the impact of BWCs on use of force among officers assigned to those units.
Police use of force
Although the authority to use force is a defining feature of police work (Bittner 1970; Klockars 1996), its actual use is statistically rare: There were 53.5 million police-citizen contacts in the United States in 2015, and force was threatened or used in only 2% of those encounters, totaling about 2900 uses of force per day (Davis et al. 2018). The legal standard for evaluating police use of force, including deadly force, comes from the United States Supreme Court’s ruling in Graham v. Connor (1989), which defines appropriate force as that which a reasonable officer would have perceived as necessary, in the moment, given the totality of the circumstances. This acknowledges the potential for harm inherent in policing and the split-second decision-making that can affect officer perceptions. However, the standard has been criticized for allowing too much subjectivity in evaluating use of force (Alpert and Smith 1994; Terrill 2016) and for failing to provide mechanisms to ensure officers do not use unnecessary or excessive force (Mears et al. 2017; White 2016). Use of force by police and the Graham standard have come under scrutiny amidst a backdrop of concerns about race- and class-based disparities, tensions following high-profile police shootings, and a perceived lack of accountability for abuse of police authority. These events have led to call for a reform of police training and management, with a focus on use of force (e.g., President’s Task Force on Twenty-First Century Policing 2015).
Efforts to study police use of force have been complicated by several factors, including variations in use of force guidelines (Terrill and Paoline 2012), inconsistency in conceptualization of key measures (Garner et al. 2002; Klahm et al. 2014), and the availability and quality of the data (White 2016). As such, this research has generally produced mixed results (Nix et al. 2017). Still, it is clear that an exceedingly small portion of police-citizen contacts devolve into violence, and force that is used is most often low in severity (Adams 1999; Bayley and Garofalo 1989; Friedrich 1980; Fyfe 2002; Garner and Maxwell 2002; Hickman et al. 2008; Reiss 1971; Worden 1995).
Correlates of police use of force
Numerous studies have examined officer- and suspect-level predictors of force. Researchers have consistently reported that the majority of suspects who are shot at by the police presented an imminent danger to the officer at the time of the shooting (Fyfe 1980, 1981; Klinger 2004). Similarly, research shows that suspect resistance is the best predictor of less-lethal force (Mulvey and White 2014; Terrill and Mastrofski 2002). Studies also show that more educated and experienced officers use less force (Garner and Maxwell 2002; International Association of Chiefs of Police 2002; McElvain and Kposowa 2008; Paoline and Terrill 2007; Schuck and Rabe-Hemp 2007). One recent study found that military veterans in the Dallas Police Department were more likely to be involved in a shooting, with combat-experienced veterans almost three times more likely (Reingle Gonzalez et al. 2019). Beyond these variables, few officer-level predictors have shown consistent relationships with officer force behaviors. Some studies have found women use less force (Brandl et al. 2001; Garner et al. 1995) while others have found no differences (Lersch 1998; Paoline and Terrill 2004; Terrill and Mastrofski 2002), and officer race has not been found to affect use of force (Friedrich 1980; Terrill and Mastrofski 2002).
Scholars have also focused on organizational and ecological correlates of use of force (Friedrich 1980; Sun and Payne 2004). Environmentally, community characteristics such as racial composition, arrest rates, economic inequality, and various measures of public violence have been tied to police use of force rates. Levels of community violence, in particular, are often positively related to rates of police use of force (Kania and Mackey 1977; Liska and Yu 1992; Matulia 1985; Terrill and Reisig 2003). Organizationally, informal organizational culture (Terrill et al. 2003) and administrative policy (Fyfe 1979; Geller and Scott 1992) are perhaps the most important factors guiding and controlling police use of force, whereas vague or unenforced policies can lead to higher rates of deadly force (White 2001). Shjarback and White (2016) also found a link between departmental commitment to education (e.g., college credit requirements for applicants) and lower rates of police-citizen violence.
Police use of force and specialized units
Specialty units were implemented in American policing during the early twentieth century as part of a larger movement to professionalize the police through bureaucratic principles (White 2007). In order to increase the efficiency of police to address crime, agencies specially trained officers to focus on specific problems, such as vagrancy or gang violence. While officers assigned to general patrol tend to handle an array of non-criminal activities where the likelihood of using force is greatly reduced (Banton 1964; Bittner 1967; Manning 1978; Mastrofski 1983), characteristics of specialty units may increase the likelihood that these officers use force against citizens (Gaub et al. 2020). For example, many specialty units are designated to respond only to high-risk calls for service, or to target violent places and people. Others are encouraged to make a large volume of arrests. Each of these factors may generate the use of aggressive enforcement tactics by officers, and greater resistance from suspects (see, e.g., Hickman et al. 2008). Additionally, Skolnick and Fyfe (1993) linked specialized units with a greater risk for extra-legal force because of competition among units, less supervision, and the potential for the emergence of a strong subculture.
With few exceptions, the research on police use of force has focused on either department-level trends or rates of force among patrol officers; only four studies have systematically examined the use of force by specialty units. Campbell et al. (1998) and Hickey and Hoffman (2003) studied dog bites in K9 units within single agencies, finding 35–45% and 14% (respectively) of suspect apprehensions resulted in a bite. Williams and Westall (2003) hypothesized that SWAT-certified patrol officers would use more force during non-SWAT calls for service than general patrol officers, but the authors found no difference between the two groups. Brandl and colleagues (Brandl et al. 2001) studied officer assignment specialization more broadly and found no effect on citizen complaints of excessive force. All four studies have limitations for drawing conclusions about differential force rates among units: The first three focus on one specialty unit, and the last examines allegations of excessive force, not actual behaviors. Additionally, only two (Brandl et al. 2001; Williams and Westall 2003) compared specialty units with patrol.
Police body-worn cameras and use of force
BWCs have diffused rapidly in American policing, in large part because of the potential for cameras to reduce officers’ use of force. Nineteen studies have tested the effects of BWCs on police use of force, and roughly half (11) documented substantial or statistically significant declines following camera deployment (White et al. 2019). The Rialto (CA) Police Department reported a substantial decrease in use of force following the adoption of BWCs, and that decline persisted for several years (Ariel et al. 2015; Sutherland et al. 2017). Other studies also reported significant declines in use of force post-BWC implementation (e.g., Braga et al. 2018b; Henstock and Ariel 2017; Jennings et al. 2017; Jennings et al. 2015; White et al. 2018a). The other half, however, found no significant change in officer use of force following BWC implementation (Ariel 2017; Ariel et al. 2015; Braga et al. 2018a; Edmonton Police Service 2015; Headley et al. 2017; Henstock and Ariel 2017; Yokum et al. 2017). In particular, when findings from a study with the Washington (DC) Metropolitan Police Department were released (Yokum et al. 2017), many questioned whether the “BWC hype” was unfounded (Ripley and Williams 2017). These mixed findings led Lum and colleagues (2019) to conclude that “it may be fair to say…that BWCs have not produced dramatic changes in police behavior, for better or worse […and] perhaps anticipated effects from BWCs have been overestimated” (pp. 19–20).
Several factors could account for these mixed findings, including the degree of discretion afforded to officers in BWC activation (Ariel et al. 2016), aspects of BWC implementation (White et al. 2018b), and the state of the department pre-BWC deployment (Malm 2019; White 2019). It is also possible the effect of BWCs could vary across work groups within an agency. Much like the body of research on use of force generally, no BWC study has examined group-level differences in use of force within an agency. It could be that BWCs lead to decreases (or increases) in use of force among some groups but not others. These group-level differences would be masked by agency-level analyses. Given the high costs associated with deploying BWCs (White 2014) and more general concerns about the use of force and its consequences, it is important for researchers to delve more deeply into the dynamics between use of force, BWC deployment, and officer work groups. Thus, these mixed findings suggest that a more nuanced investigation into the impact of BWCs on use of force is warranted, which is the focus of the current study.
Data and methods
The current analysis is part of a larger 6-month randomized controlled trial (RCT) evaluation of BWC implementation in the Tempe (AZ) Police Department.Footnote 1 Officers in the patrol division (composed of both patrol and several specialty units [bicycle, K9, tactical response,Footnote 2 gang, mounted, and traffic]) were randomly assigned to either the treatment group (phase 1, N = 101) or control group (phase 2, N = 99). Phase 1 officers received BWCs in November 2015, and phase 2 officers received BWCs 6 months later in May 2016; thus, the RCT period is the 6-month interval in which phase 1 (treatment) officers had BWCs and phase 2 (control) officers did not. While BWC deployment was randomized, specialty unit assignment was not randomized; thus, assignment to specialty or patrol unit is quasi-experimental. Of the 200 study officers, 87 were assigned to a specialty unit for at least 1 month during the study period. In any given month, between 48 and 64 officers were assigned to a specialty unit. On average, specialty unit officers were assigned to any specialized unit for 31.3 of the 48 study months. It was rare for officers to either move (a) from a specialty unit, into patrol, and then back into a specialty unit again, or (b) from one specialty unit to another (N = 4 for each).
Tempe, Arizona, is located just southeast of downtown Phoenix and is home to Arizona State University (ASU)’s main campus. Tempe has a population of approximately 180,000 permanent residents and is predominantly White (72%) and Latinx (21%). The median household income in 2016 was $50,474, with about one-fifth of residents living under the poverty line. Violent and property crimes were both well above the national average in 2016 (504.9 and 4558.5 per 100,000 residents, respectively). The police department is medium-sized, employing approximately 330 sworn officers. The main patrol division is divided into four areas: North, South, Central, and Traffic. The department has several specialized police units; those participating in this study include the bike unit, K9 unit, Tactical Response Unit (TRU), gang unit, mounted unit, and traffic unit. Due to study parameters, only specialty units considered part of the patrol division were included in the RCT evaluation, and thus are included here. Other specialty units, such as the Criminal Apprehension and Surveillance Team or undercover narcotics and vice units, never received BWCs due to the nature of their job assignment; the Criminal Investigations Bureau received theirs long after the RCT.
We collected administrative data on use of force, calls for service, and weekly assignment rosters for a 48-month period beginning 1 year prior to the BWC RCT (November 1, 2014) and ending October 31, 2018. We constructed monthly force rates for each officer in the study, per 1000 calls for service ([monthly number of force incidents/number of calls for service] × 1000). Additionally, we separated each officer’s monthly assignment in a specialty (1) or patrol (0) unit. Using these data, we aggregated force rates by unit (patrol vs. specialty), experimental group (phase 1 vs. phase 2), and combinations therein (see Table 1).
First, we examine use of force trends during the study period by both unit type (specialty vs. patrol) and research group (phase 1 vs. phase 2), using independent (between-group) and dependent (within-group) t tests of means. Next, we separate each unit type by research group, permitting comparison among the four officer groups: phase 1 specialty unit, phase 2 specialty unit, phase 1 patrol, and phase 2 patrol. Rates are compared using six-month periods to be consistent with the six-month RCT period. Finally, we conducted autoregressive integrated moving average (ARIMA) analyses to ascertain whether use of force rates across any of those four officer groups changed significantly following BWC deployment. ARIMA is a quasi-experimental design that allows for comparison of pre-intervention and post-intervention (i.e., pre- and post-BWC deployment) values of an outcome, in this case, use of force. In this study, the technique determines whether the trend in use of force changes following BWC deployment. ARIMA overcomes several threats to internal validity and violations of the independence assumption (e.g., serial correlation; McDowall and McCleary 2014; McDowall et al. 1980).
Unit and research group assignment
Figure 1 displays the monthly force rates among officers assigned to specialty and patrol units, standardized by call activity. Officers assigned to specialty units consistently use force more frequently than officers assigned to patrol. The first two columns of Table 1 show average force rates for each officer work group, broken down into 6-month intervals. Specifically, force rates for officers in specialty units range from a low of 1.81 to a high of 4.33 times the rates for patrol unit officers. There are no statistically significant within-group differences, but between-group differences (specialty compared with patrol) are significant at every time-block (p < 0.05). Intuitively, this is logical, since specialty unit officers are summoned to more contentious and potentially violent calls than patrol officers (Gaub et al. 2020). However, the only studies to empirically test this assertion failed to find statistically significant differences in use of force between patrol and specialty unit officers (Brandl et al. 2001; Williams and Westall 2003). Our study contradicts these findings.
The third and fourth columns of Table 1 show the standardized rates of force for all 200 study officers separated by BWC research group assignment. Using t tests of means, we find that there are no statistically significant within-group or between-group differences. At the aggregate group level, BWCs had no impact on the rates of use of force, a finding that is consistent with more recent studies of BWCs and use of force (White et al. 2019).
We next tested whether the overall trends in use of force were masked by group-level change, which is a possibility for two reasons. First, specialty unit officers comprise a relatively small group: Only 48–64 officers per month are members of a specialty unit in this agency. Additionally, specialty units are called to incidents that typically receive greater scrutiny; thus, these officers may respond differently to being filmed compared with those working general patrol. To test this hypothesis, we separate officers by both research group assignment and unit assignment (see remaining columns in Table 1) and compared rates among the groups over time.
Notably, phase 1 specialty unit officers used significantly less force than phase 2 specialty unit officers for the first 2 years after BWCs were deployed to phase 1 officers (November 2015–October 2017), with no corresponding difference in patrol. This difference diminished for the last year of the study period (November 2017–October 2018). Additional analysis showed that no single specialty unit was driving this finding in terms of number of phase 1 or 2 officers; in other words, no single unit has far fewer phase 1 officers, or far more phase 2 officers, compared with others. This eliminates the possibility that the proportion of phase 1 and 2 officers in any specific specialty unit was driving this finding. There are, however, several alternative explanations for this finding. For example, the finding could be driven by specific officers, opportunity to engage in force, or a combination thereof. Officers in specialty units have, by virtue of their mission and assignment, greater opportunity to engage in use of force. As such, when they are moved back into patrol, the opportunity or inclination to respond with force is removed or mitigated. In short, there are noteworthy but perplexing group differences within specialty units as a whole.
Time series analysis
We identified the best-fitting model for all phase 1 officers (2,1,0) and all phase 2 officers (0,1,1), as well as each subgroup: phase 1 patrol (2,1,0), phase 1 specialty (1,0,1), phase 2 patrol (0,1,1), and phase 2 specialty (0,1,1) (see Table 2). We then conducted interrupted time series analysis with each officer group, testing BWC impact with varying onsets (abrupt, gradual) and duration (temporary, permanent), starting in November 2015 for phase 1 officer groups and May 2016 for phase 2 officer groups. There was only one statistically significant finding: a gradual, permanent decline in use of force among phase 2 specialty officers starting in February 2017 (9-month delayed onset after BWC deployment) and persisting through the end of the study period (see also Fig. 2). The introduction of BWCs was not associated with change in force rates for any other officer group.
While there is a robust literature analyzing correlates of police coercive behavior and a growing body of research on BWCs, scholars in both fields have traveled the same path and left the same question unanswered: are there unit-level differences in use of force? Our study bridges this gap in both areas of research, and we have two primary findings.
First, use of force rates for officers assigned to specialty units were nearly four times greater than those of officers assigned to general patrol units, and this finding was statistically significant over the entire study period. We account for unit-level variation in call activity and unit assignment changes. Still, Tempe officers in specialty units use force far more frequently than officers in patrol units. To our knowledge, this is the first study to identify such disparate rates of force among officer work groups. Second, we find that the implementation of BWCs had no impact on use of force at the department level. This is true cross-sectionally, looking at each time-block using t tests of means, and longitudinally using ARIMA, though we see a slightly different story at the unit level.
Given the serious implications of police use of force, it is critical that we have a thorough understanding of not only department-level trends, but also the prevalence and distribution of force within a department. Only two studies have addressed use of force in specialized police units: One focused on one specialty unit (SWAT; Williams and Westall 2003) and the other examined citizen complaints of excessive force rather than actual use of force behaviors (Brandl et al. 2001). By comparison, our study is a rigorous assessment of all officer use of force over a 4-year period comparing multiple specialty units with general patrol, and we find that officers in specialty units consistently use force at a higher rate than patrol officers. This suggests that significantly more attention should be paid to the behaviors of police specialty units. Given the current focus placed on police use of force and public unrest concerning its perceived excessive use against minorities, the dearth of research addressing coercive behaviors used by the specialty units who respond to high-risk calls for service is surprising.
The mission of specialty units and the nature of their work provide intuitive explanations for this finding. By definition, specialty units have a narrowly defined mission related to handling particular crime or offender types (e.g., traffic, gangs), geographic areas (e.g., bikes, mounted patrol), or tactical needs (e.g., TRU, K9). Their mission is grounded almost exclusively in law enforcement and crime control. Conversely, patrol officers handle a wide breadth of tasks, ranging from traffic control to community policing to responding to calls for service. Specialty units also operate in high-risk places, are summoned to high-risk calls, and interact primarily with high-risk individuals. This combination of factors can make verbal de-escalation difficult, perhaps increasing the likelihood that force is required. For example, mounted and bike officers in Tempe experience a significantly high call volume and address various problems in quick succession, all confined within in a small geographic area where alcohol and large numbers of people mix and mingle. According to Todak and White (2019), police officers find it most difficult to verbally de-escalate intoxicated individuals because they are less responsive to rational arguments and other non-physical persuasion tactics. Similarly, K9, gang, and TRU officers are more likely to respond to calls involving high-risk suspects and requiring the use of more advanced tactics and weaponry.
The substantial difference in rates of force among officers assigned to specialty units compared with general patrol also poses several theoretical questions. Is there group-level variation in the causes and correlates of use of force? Are some predictors more salient for one group of units than others? Several possibilities emerge. The riskier environment in which specialty unit officers work could indicate that environmental factors play a larger role in predicting use of force. Alternatively, the fact that specialty units interact with riskier people could suggest that suspect characteristics or behaviors have a greater influence on officer responses. Or perhaps, it is a confluence of the two—and the reliance on specialty units at high-risk calls—which would indicate that situational factors have the greatest effect. There may also be subcultural influences at the unit level at play, as suggested by Sherman’s “rotten barrels” hypothesis (see Skolnick and Fyfe 1993). Future research should look to answer these questions.
The idea that officers in police specialty units are disproportionately high users of force also has important policy implications. The selection process for these units typically involves a physical and/or written examination, as well as recommendations from supervisors or peers. Yet, the increased risk of the specialized job assignment also indicates that officers should be proficient in verbal de-escalation skills, and to have what Muir (1977, p. 50) called “passion and perspective.” Muir argued that the professional policeman possesses both a strong sense of empathy (i.e., perspective, which he described as an understanding of “the dignity and tragedy of the human condition”) and moral acceptance of the need to sometimes use force (i.e., passion; p. 50). Though Muir’s focus was on officers assigned to patrol, the passion and perspective traits are as important—arguably more so—for officers assigned to specialty units. Our findings suggest the importance of selecting individuals for specialty units who possess traits and skills that can reduce the chances of negative outcomes.
To that point, officers in these units should receive more training opportunities related to the challenges they face, such as handling mentally ill individuals (e.g., Crisis Intervention Training [CIT]) and verbal violence de-escalation. These units may also require a more nuanced use of force policy that acknowledges the nature of their work. BWC-specific training and policy should also account for the unique needs of specialty units, including possible adjustments in terms of de/activation, public release of footage, and citizen notification (Gaub et al. 2020).
Finally, the risk of unnecessary and excessive force is also likely higher in specialty units. To be clear, our analysis did not address justified vs. unjustified use of force, but it is simply a function of frequency; by virtue of officers in specialty units using more force, there is more opportunity for unnecessary/excessive force. Departments must ensure that adequate supervision and accountability mechanisms are in place for those units, as prior research suggests that organizational specialization may be tied to increased risk of misconduct (Kappeler et al. 1998).
The lack of department-wide change in use of force following BWC deployment may be explained by two key factors. First, the Tempe Police Department closely adhered to the process and principles outlined in the U.S. Department of Justice National BWC Toolkit’s Implementation Checklist and engaged in an 18-month planning process before deploying a single camera (Bureau of Justice Assistance 2016; White et al. 2018b). The department formed a working group composed of department representatives (including line officers) and other stakeholders such as city and legal representatives. The working group invited external stakeholders to their meetings and solicited their input on relevant policy development issues. Additionally, command staff attended every roll call briefing to explain the program and evaluation, introduce the research team, and answer questions from officers.
Second, TPD is widely regarded as high-functioning, progressive, and professional, as evidenced by their commitment to evidence-based policing and the election of Chief Sylvia Moir as President of the Police Executive Research Forum in late 2019 (White et al. 2018b; Police Executive Research Forum 2019). White and Malm (2020 pp. 37–38; see also Malm 2019) note that large reductions in use of force are unlikely to occur in such departments:
[P]rofessional organizations with restrictively high selection standards, robust training, effective supervision, and proper accountability systems do not experience those same negative outcomes because they were never in a bad place from which to recover…There were no large reductions in use of force and citizen complaints because there did not need to be—officers were using appropriate levels of force already.
Thus, it is unsurprising that BWCs would have little effect on use of force in Tempe, as has been the case with similarly professional or reformed departments such as Denver (Ariel 2017), Milwaukee (Peterson et al. 2018), and Washington, DC (Yokum et al. 2017). And while Sutherland and colleagues (2017) found the significant decline in use of force in the Rialto (CA) Police Department persisted over several years, Koslicki and colleagues (2020) found that use of force rates in an unnamed police department significantly yet gradually returned to pre-BWC levels after a nonsignificant drop at the point of implementation. Our study provides a counterpoint to both, strongly suggesting that there is no single BWC experience (see also White and Malm 2020).
Unlike the findings at the department level, our unit-level analyses are less clear. Given that there is no substantial, significant change in force rates as the BWC RCT begins, it is difficult to account for the significant decline among phase 2 specialty unit officers 9 months after they received BWCs. Regardless of the reason, there is clearly the potential for BWCs to have differential effects on officer use of force, which should be explored further in future research.
Like any study, ours has limitations. First, official measures of police use of force are imperfect. Malm (2019, p. 14) argues that scholars must “start to agree on some universal metrics,” which would include responding to departmental variation in definitions of reportable uses of force (i.e., the threshold issue). Lum et al. (2019) echo this concern, questioning whether official records offer the most appropriate measures for examining police use of force behaviors. Future research should begin to parse out the nuances of the BWC use of force relationship and examine what the observed differences mean in practice. Researchers could, for example, focus on differences at the unit level (as we do here) or use alternative forms of data such as systematic social observation or analysis of BWC footage itself.
Our study also sampled specialty units according to two criteria: (1) the unit operated in the Tempe Police Department and (2) the unit received BWCs as part of the research study. There were additional units in the police department that did not receive cameras during the study, such as the Criminal Apprehension and Surveillance Team or the detectives of the Criminal Investigations Bureau, and were therefore not included here. The implementation of BWCs within other specialty units or in other police departments may produce different findings; future research should test the generalizability of our findings, especially in jurisdictions with specialty units varying in type and size.
Finally, our data did not provide context regarding the situational factors of calls. Prior research has shown these factors to be particularly salient for explaining force incidence (Fyfe 1980, 1981; Klinger 2004; Mulvey and White 2014; Terrill and Mastrofski 2002), but these data were not available. Future research should investigate how situational factors—such as call type, suspect resistance, and arrest information—differentially affect use of force at the unit level. This would be a particularly fruitful avenue of research using BWC footage as the data source.
Our study makes two contributions to the ongoing dialog regarding use of force and police BWCs. We demonstrate that, in the Tempe Police Department, (1) use of force is concentrated among specialty units, and (2) BWCs have the potential to generate a differential effect on use of force at the group level. Both of these issues warrant more investigation. As communities demand greater transparency and accountability, agencies have quickly deployed BWCs. Yet inclusion of specialty units in BWC programs has been inconsistent, often with the most force-prone units (such as criminal apprehension, SWAT, and anti-crime units) being excluded from wearing cameras (Gaub et al. 2020). If increased accountability and controlling officer force are key goals of a BWC program, perhaps agencies should rethink this strategy.
The larger evaluation encompassed a number of other components (see Gaub et al. 2016, 2020; White et al. 2018b; Todak et al. 2018). The randomization protocol called for all officers below the rank of lieutenant (including sergeants, officers, and some designated as detectives) who were assigned to the patrol division (N = 200) to be randomly assigned to receive a BWC during either phase 1 (November 2015) or phase 2 (May 2016). The department had already planned to use a phased approach to deploying BWCs, so randomization permitted an experimental design.
The Tactical Response Unit (TRU) is the full-time SWAT unit; several officers in other assignments are SWAT-certified and respond to calls with TRU when a larger response is needed.
Adams, K. (1999). What we know about police use of force. Use of force by police: overview of national and local data (pp. 1–14).
Alpert, G. P., & Smith, W. C. (1994). How reasonable is the reasonable man: police and excessive force. The Journal of Criminal Law and Criminology (1973-), 85(2), 481–501.
Ariel, B. (2017). Police body cameras in large police departments. The Journal of Criminal Law and Criminology, 106(4), 729–768.
Ariel, B., Farrar, W. A., & Sutherland, A. (2015). The effect of police body-worn cameras on use of force and citizens’ complaints against the police: a randomized controlled trial. Journal of Quantitative Criminology, 31(3), 1–27.
Ariel, B., Sutherland, A., Henstock, D., Young, J., Drover, P., Sykes, J., Megicks, S., & Henderson, R. (2016). Report: increases in police use of force in the presence of body-worn cameras are driven by officer discretion: a protocol-based subgroup analysis of ten randomized experiments. Journal of Experimental Criminology, 12(3), 453–463.
Banton, M. (1964). The policeman in the community. New York: Basic Books.
Bayley, D. H., & Garofalo, J. (1989). The management of violence by police patrol officers. Criminology, 27(1), 1–26.
Bittner, E. (1967). The police on skid-row: a study of peace keeping. American Sociological Review, 32(5), 699–715.
Bittner, E. (1970). The functions of the police in modern society. Bethesda: National Institute of Mental Health.
Braga, A. A., Barao, L., McDevitt, J., & Zimmerman, G. (2018a). The impact of body-worn cameras on complaints against officers and officer use of force incident reports: preliminary evaluation findings. Boston: School of Criminology and Criminal Justice, Northeastern University.
Braga, A. A., Sousa, W., Coldren, J. R., & Rodriguez, D. (2018b). The effects of body worn cameras on police activity and police-citizen encounters: a randomized controlled trial. The Journal of Criminal Law and Criminology, 108(3), 511–538.
Brandl, S. G., Stroshine, M. S., & Frank, J. (2001). Who are the complaint-prone officers?: an examination of the relationship between police officers’ attributes, arrest activity, assignment, and citizens’ complaints about excessive force. Journal of Criminal Justice, 29(6), 521–529.
Bureau of Justice Assistance. (2016). Body-Worn Camera Toolkit. https://www.bja.gov/bwc/.
Campbell, A., Berk, R. A., & Fyfe, J. J. (1998). Deployment of violence: the Los Angeles Police Department’s use of dogs. Evaluation Review, 22(4), 535–561.
Davis, E., Whyde, A, & Langton, L. (2018). Contacts between police and the public, 2015. NCJ 251145. Bureau of Justice Statistics.
Edmonton Police Service. (2015). Body worn video: Considering the evidence. In Final report of the Edmonton police service body worn video pilot project. Edmonton: Edmonton Police Service.
Friedrich, R. J. (1980). Police use of force: individuals, situations, and organizations. The Annals of the American Academy of Political and Social Science, 452(1), 82–97.
Fyfe, J. J. (1979). Administrative interventions on police shooting discretion: an empirical examination. Journal of Criminal Justice, 7(4), 309–324.
Fyfe, J. J. (1980). Geographic correlates of police shooting: a microanalysis. Journal of Research in Crime and Delinquency, 17(1), 101–113.
Fyfe, J. J. (1981). Observations on police deadly force. Crime & Delinquency, 27(3), 376–389.
Fyfe, J. J. (1988). Police use of deadly force: research and reform. Justice Quarterly, 5(2), 165–205.
Fyfe, J. J. (2002). Too many missing cases: holes in our knowledge about police use of force. Justice Research and Policy, 4(1), 87–102.
Garner, J. H., & Maxwell, C. D. (2002). Understanding the use of force by and against the police in six jurisdictions. Washington, DC: National Institute of Justice.
Garner, J. H., Schade, T., Hepburn, J., & Buchanan, J. (1995). Measuring the continuum of force used by and against the police. Criminal Justice Review, 20(2), 146–168.
Garner, J. H., Maxwell, C. D., & Heraux, C. G. (2002). Characteristics associated with the prevalence and severity of force used by the police. Justice Quarterly, 19(4), 705–746.
Gaub, J. E., Choate, D. E., Todak, N., Katz, C. M., & White, M. D. (2016). Officer perceptions of body-worn cameras before and after deployment: a study of three departments. Police Quarterly, 19(3), 275–302.
Gaub, J. E., Todak, N., & White, M. D. (2020). One size doesn’t fit all: the deployment of police body-worn cameras to specialty units. International Criminal Justice Review, 30(2), 136–155.
Geller, W. A., & Scott, M. (1992). Deadly force: what we know: a practitioner’s desk reference on police-involved shootings. Washington, DC: Police Executive Research Forum.
Headley, A. M., Guerette, R. T., & Shariati, A. (2017). A field experiment of the impact of body-worn cameras (BWCs) on police officer behavior and perceptions. Journal of Criminal Justice, 53, 102–109.
Henstock, D., & Ariel, B. (2017). Testing the effects of police body-worn cameras on use of force during arrests: a randomised controlled trial in a large British police force. European Journal of Criminology, 14(6), 720–750.
Hickey, E. R., & Hoffman, P. B. (2003). To bite or not to bite: canine apprehensions in a large, suburban police department. Journal of Criminal Justice, 31(2), 147–154.
Hickman, M. J., Piquero, A. R., & Garner, J. H. (2008). Toward a national estimate of police use of nonlethal force. Criminology & Public Policy, 7(4), 563–604.
International Association of Chiefs of Police. (2002). Police use of force in America. Alexandria: International Association of Chiefs of Police.
Jennings, W. G., Lynch, M. D., & Fridell, L. (2015). Evaluating the impact of police officer body-worn cameras (BWCs) on response-to-resistance and serious external complaints: evidence from the Orlando Police Department (OPD) experience utilizing a randomized controlled experiment. Journal of Criminal Justice, 43(6), 480–486.
Jennings, W. G., Fridell, L. A., Lynch, M., Jetelina, K. K., Reingle, J. M., & Gonzalez. (2017). A quasi-experimental evaluation of the effects of police body-worn cameras (BWCs) on response-to-resistance in a large metropolitan police department. Deviant Behavior, 38(11), 1332–1339.
Kania, R. R. E., & Mackey, W. C. (1977). Police violence as a function of community characteristics. Criminology, 15(1), 27–48.
Kappeler, V. E., Sluder, R. D., & Alpert, G. P. (1998). Forces of deviance: understanding the dark side of policing (2nd ed.). Long Grove: Waveland Press.
Klahm, C. F., Frank, J., & Liederbach, J. (2014). Understanding police use of force. Policing: An International Journal of Police Strategies & Management, 37(3), 558–587.
Klinger, D. (2004). Environment and organization: reviving a perspective on the police. The Annals of the American Academy of Political and Social Science, 593(1), 119–136.
Klinger, D. (2008). On the importance of sound measures of forceful police actions. Criminology & Public Policy, 7(4), 605–617.
Klockars, C. B. (1996). A theory of excessive force and its control. In W. A. Geller & H. Toch (Eds.), Police violence: understanding and controlling police abuse of force (pp. 1–22). New Haven: Yale University Press.
Koslicki, W. M., Makin, D. A., & Willits, D. (2020). When no one is watching: evaluating the impact of body-worn cameras on use of force incidents. Policing and Society.
Lersch, K. M. (1998). Exploring gender differences in citizen allegations of misconduct. Women & Criminal Justice, 9(4), 69–79.
Liska, A. E., & Yu, J. (1992). Specifying and testing the threat hypothesis: police use of deadly force. In A. E. Liska (Ed.), Social threat and social control (pp. 53–68). Albany: State University of New York Press.
Lum, C., Stoltz, M., Koper, C. S., & Amber Scherer, J. (2019). The research on body-worn cameras: what we know, what we need to know. Criminology & Public Policy, 18(1), 93–118.
Malm, A. (2019). The promise of body-worn cameras. Criminology & Public Policy, 18(1), 119–130.
Manning, P. K. (1978). The police: mandate, strategies, and appearances. In P. K. Manning & J. Van Maanen (Eds.), Policing: A view from the street (pp. 7–31). New York: Random House.
Mastrofski, S. D. (1983). The police and noncrime services. In G. P. Whitaker & C. Phillips (Eds.), Evaluating the performance of criminal justice agencies. Beverly Hills: Sage Publications.
Matulia, K. R. (1985). A balance of forces (2nd ed.). Gaithersburg: International Association of Chiefs of Police.
McDowall, D., & McCleary, R. (2014). Interrupted time series models. In G. Bruinsma & D. Weisburd (Eds.), In Encyclopedia of criminology and criminal justice (Vol. 2). New York: Springer.
McDowall, D., McCleary, R., Meidinger, E. E., & Hay, R. A. (1980). Interrupted time series analysis. Newbury Park: Sage.
McElvain, J. P., & Kposowa, A. J. (2008). Police officer characteristics and the likelihood of using deadly force. Criminal Justice and Behavior, 35(4), 505–521.
Mears, D. P., Craig, M. O., Stewart, E. A., & Warren, P. Y. (2017). Thinking fast, not slow: how cognitive biases may contribute to racial disparities in the use of force in police-citizen encounters. Journal of Criminal Justice, 53, 12–24.
Muir, W. K. (1977). Police: streetcorner politicians. Chicago: University of Chicago Press.
Mulvey, P., & White, M. D. (2014). The potential for violence in arrests of persons with mental illness. Policing: An International Journal of Police Strategies & Management, 37(2), 404–419.
Nix, J., 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.
Paoline, E. A., & Terrill, W. (2004). Women police officers and the use of coercion. Women & Criminal Justice, 15(3–4), 97–119.
Paoline, E. A., & Terrill, W. (2007). Police education, experience, and the use of force. Criminal Justice and Behavior, 34(2), 179–196.
Peterson, B. E., Yu, L., La Vigne, N., & Lawrence, D. S. (2018). The Milwaukee Police Department’s body-worn camera program. Washington, D.C.: Urban Institute.
Police Executive Research Forum. (2019). PERF welcomes new board members. Subject to Debate, 33(2) https://www.policeforum.org/assets/December2019Debate.pdf.
President’s Task Force on 21st Century Policing. (2015). Final report of the President’s task force on 21st century policing. Washington, DC: Office of Community Oriented Policing Services.
Reaves, B. A. (2015). Local police departments, 2013: personnel, policies, and practices. Washington, DC: Bureau of Justice Statistics.
Reingle Gonzalez, J. M., Bishopp, S. A., Jetelina, K. K., Paddock, E., Gabriel, K. P., & Brad Cannell, M. (2019). Does military veteran status and deployment history impact officer involved shootings? A case-control study. Journal of Public Health, 19.
Reiss, A. J. (1971). The police and the public. New Haven: Yale University Press.
Ripley, A., & Williams, T. (2017). Body cameras have little effect on police behavior, study says. The New York Times, 20, 2017.
Schuck, A. M., & Rabe-Hemp, C. (2007). Women police: the use of force by and against female officers. Women & Criminal Justice, 16(4), 91–117.
Shjarback, J. A., & White, M. D. (2016). Departmental professionalism and its impact on indicators of violence in police–citizen encounters. Police Quarterly, 19(1), 32–62.
Skolnick, J. H., & Fyfe, J. J. (1993). Above the law: police and the excessive use of force. New York: The Free Press.
Sun, I. Y., & Payne, B. K. (2004). Racial differences in resolving conflicts: a comparison between Black and White police officers. Crime & Delinquency, 50(4), 516–541.
Sutherland, A., Ariel, B., Farrar, W., & De Anda, R. (2017). Post-experimental follow-ups—fade-out versus persistence effects: the rialto police body-worn camera experiment four years on. Journal of Criminal Justice, 53, 110–116.
Terrill, W. (2016). Deadly force: to shoot or not to shoot. Criminology & Public Policy, 15(2), 491–496.
Terrill, W., & Mastrofski, S. D. (2002). Situational and officer-based determinants of police coercion. Justice Quarterly, 19(2), 215–248.
Terrill, W., & Paoline, E. A. (2012). Examining less lethal force policy and the force continuum: results from a national use-of-force Study. Police Quarterly, 16(1), 38–65.
Terrill, W., & Reisig, M. D. (2003). Neighborhood context and police use of force. Journal of Research in Crime and Delinquency, 40(3), 291–321.
Terrill, W., Paoline, E. A., & Manning, P. K. (2003). Police culture and coercion. Criminology, 41(4), 1003–1034.
The National Advisory Commission on Civil Disorders. (1968). Report of the National Advisory Commission on civil disorders. Washington, DC: US Government Printing Office.
Todak, N., & White, M. D. (2019). Expert officer perceptions of de-escalation in policing. Policing: An International Journal, 42(5), 832–846.
Todak, N., Gaub, J. E., & White, M. D. (2018). The importance of external stakeholders for police body-worn camera diffusion. Policing: An International Journal, 41(4), 448–464.
White, M. D. (2001). Controlling police decisions to use deadly force: reexamining the importance of administrative policy. Crime & Delinquency, 47(1), 131–151.
White, M. D. (2007). Current issues and controversies in policing. Boston: Allyn and Bacon/Pearson.
White, M. D. (2014). Police officer body-worn cameras: assessing the evidence. Washington, DC: U.S. Department of Justice, Office of Justice Programs.
White, M. D. (2016). Transactional encounters, crisis-driven reform, and the potential for a national police deadly force database. Criminology & Public Policy, 15(1), 223–235.
White, M. D. (2019). Translating the story on body-worn cameras. Criminology & Public Policy, 18(1), 89–91.
White, M. D., & Fradella, H. F. (2016). Stop and frisk: the use and abuse of a controversial policing tactic. New York: New York University Press.
White, M. D., & Klinger, D. (2012). Contagious fire? An empirical assessment of the problem of multi-shooter, multi-shot deadly force incidents in police work. Crime & Delinquency, 58(2), 196–221.
White, M. D., & Malm, A. (2020). Cops, cameras, and crisis: the emergence of body-worn cameras in 21st century policing. New York: New York University Press.
White, M. D., Gaub, J. E., & Todak, N. (2018a). Exploring the potential for body-worn cameras to reduce violence in police-citizen encounters. Policing: A Journal of Policy and Practice, 12(1), 66–76.
White, M. D., Todak, N., & Gaub, J. E. (2018b). Examining body-worn camera integration and acceptance among police officers, citizens, and external stakeholders. Criminology & Public Policy, 17(3), 649–677.
White, M. D., Gaub, J. E., & Padilla, K. E. (2019). Impacts of BWCs on use of force: directory of outcomes. Body-Worn Camera Training and Technical Assistance.. http://www.bwctta.com/resources/bwc-resources/impacts-bwcs-use-force-directory-outcomes.
Williams, J. J., & Westall, D. (2003). SWAT and non-SWAT police officers and the use of force. Journal of Criminal Justice, 31(5), 469–474.
Worden, R. E. (1995). The causes of police brutality: theory and evidence on police use of force. In W. A. Geller & H. Toch (Eds.), And justice for all: understanding and controlling police abuse of force (pp. 31–60). Washington, DC: Police Executive Research Forum.
Yokum, D., Ravishankar, A., & Coppock, A. (2017). Evaluating the effects of police body-worn cameras: a randomized controlled trial. Washington, DC: The Lab @ DC, Office of the City Administrator, Executive Office of the Mayor.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Gaub, J.E., Todak, N. & White, M.D. The distribution of police use of force across patrol and specialty units: a case study in BWC impact. J Exp Criminol (2020). https://doi.org/10.1007/s11292-020-09429-8
- Body-worn cameras (BWCs)
- Use of force
- Randomized controlled trial (RCT)
- Specialty units