Citizen Perceptions Of Body-Worn Cameras: A Randomized .

8d ago
1.62 MB
57 Pages

Citizen Perceptions of Body-WornCameras: A Randomized Controlled TrialFinal ReportApril 2017Police Executive Research Forum1120 Connecticut Avenue NW, Suite 930Washington, DC 200361

Table of ContentsExecutive Summary. 4Introduction . 9Existing BWC Research . 9Impact of BWCs on Citizen Perceptions of Police . 10PERF’s 2014 BWC Recommendations. 10BWC Pilot Study – Arlington (TX) Police Department . 10Expectations of the Study . 11Location of the Study . 11Methodology . 12Study Design . 12Benefit of the Randomization Approach . 13APD Policy for BWC Use . 13Ensuring Study Fidelity. 14Determining Citizen Perceptions . 15Determining Which Citizens to Contact . 16Method of Contacting Citizens . 17Rate of Responses to Survey . 17Characteristics of Survey Respondents . 23Data on Citywide Citizen Complaints . 27Results of the Study . 27Differences in Citizen Perceptions. 27Within-Group Differences in Citizen Perceptions . 29Changes in Citizen Complaints . 31Conclusions . 32What We Learned from the Study . 32Citizen Complaints . 32Contextual Differences in Citizen Perceptions. 33Overall Impacts on Citizen Perceptions. 34Policy Implications . 34Research Limitations . 36Research Implications . 37Future Research . 38References . 402

Appendix A. Matrix of PERF’s Recommendations on the Use of BWCs (from Miller andToliver, 2014) . 43General recommendations. 43Recording protocols . 44Download and storage policies . 49Recorded data access and review . 51Training policies . 53Policy and program evaluation . 55Appendix B. Telephone Survey . 563

Executive SummaryBody-worn cameras (BWCs) have become a central topic of policing reforms within thepast few years. In the wake of recent high-profile use-of-force cases, many policedepartments accelerated their plans to implement BWCs. Conservative estimates suggestup to one-third of police departments in the U.S. are using BWCs, with that count increasingrapidly.The rapid adoption of BWCs has outpaced research into the impact that this technology hashad on policing. Most studies of BWCs to date focus on two main outcomes, namely officeruse of force and citizen complaints against officers. Research points towards significantdeclines in both of these outcomes due to BWC implementation.However, the impact of BWCs is believed to go beyond officer use of force and citizencomplaints. For instance, police officials often note that there is overall public approval ofBWCs, and that implementing a BWC program can help increase perceptions of policelegitimacy. However, this rationale had not been rigorously tested until the current study.The Police Executive Research Forum (PERF) conducted a randomized controlled trial(RCT) to assess changes in citizen perceptions due to BWC use. With support from theLaura and John Arnold Foundation (LJAF), PERF was able to conduct this RCT inpartnership with the Arlington, TX Police Department (APD). This work builds on PERF’sprior work developing an implementation guide for BWC programs and examines theimpact that a BWC program has on citizens’ opinions of the police.Expectations of the StudyPERF researchers had three expectations:1. There was an expectation that there would be fewer citizen complaints filed bycitizens who interacted with patrol officers who wore BWCs than among citizenswho interacted with patrol officers who did not wear BWCs.2. Citizens who were voluntarily involved with police (i.e. crime victims, witnesses,and people who called the police for service) were expected to have betterperceptions of police legitimacy, professionalism, and satisfaction than citizenswho were involuntarily contacted by police (i.e. criminal suspects, arrestees, andpeople pulled over for traffic stops).3. Citizens who interacted with patrol officers who wore BWCs were expected tohave significantly better perceptions of police legitimacy, satisfaction with theirinteractions, and views of police professionalism compared with citizens whointeracted with officers who did not wear BWCs.4

Study DesignPERF designed an RCT to test these expectations during APD’s BWC pilot program, whichtook place between October 2015 and March 2016. At the time of the pilot study, therewere 634 sworn officers in APD. APD is one of the largest police departments to conductan RCT to examine BWCs. Sworn officers were recruited to participate in the study fromacross the city’s four districts and from the citywide traffic enforcement unit. A total of 84volunteers were recruited for the pilot program and trained to use BWCs.Randomization among the 84 officers was done by shift. During each shift, officers wererandomly selected to wear or not wear BWCs. This means that in the course of the study, asingle officer had some shifts during which he or she wore a camera, and some shifts inwhich he or she did not. There was a 50% chance an officer would be assigned the BWCduring a single shift, akin to a simulated “coin flip.” This approach allowed the PERFresearchers to be highly confident that any differences in the outcomes of the studybetween the BWC/no-BWC groups could be attributed to the BWC condition.Data on Citywide Citizen ComplaintsThe researchers obtained data from APD on citizen complaints for the entire city during thepilot period (October 2015 – March 2016), as well as for the same six months one yearprior to the study (October 2014 – March 2015). APD also provided the complaint dataspecifically for the group of BWC-trained officers both during the pilot period and for thesame six months one year prior to the study.Determining Citizen PerceptionsTo determine citizen perceptions, PERF conducted telephone surveys of individuals whohad contact with the 84 BWC pilot officers during the six months of the pilot study. Duringthe survey, citizens were asked to recall their interactions with the APD officer on thespecific date of their contact, and were asked to rate how much they agreed or disagreedwith the callers’ statements, with 1 being “strongly disagree” and 5 being “strongly agree”(See Appendix B). The survey measured three citizen perceptions of officers: legitimacy,professionalism, and satisfaction.The pool of survey participants was created from documented encounters with APDofficers during which a formal report was generated by the police. Based on the date ofencounter and the APD officer involved, PERF was able to determine if the officer waswearing a BWC during the encounter. Additional data provided information as to whetherthe encounter was voluntary or involuntary.PERF obtained 502 completed citizen perception surveys. The respondent demographics(e.g. sex, race/ethnicity, age) closely mirrored the demographics among all reportedencounters, suggesting that the PERF respondent pool is representative of all citizensavailable to survey. Additionally, these similarities continued within the RCT between the5

treatment group (shifts when the officer was wearing a BWC) and control group (shiftswhen the officer was not wearing a BWC).Results of the StudyFollowing are the initial findings of the study, with respect to the three researchexpectations:1. Officers who were trained in BWCs during the study experienced a 38% drop incomplaints between the year prior to the pilot study and the same six months a yearlater, during the pilot study. During this time, all other APD officers experienced a4.1% increase in citizen complaints (see Figure 15). This finding was consistentwith the expectations that there would be fewer citizen complaints filed by citizenswho interacted with officers who wore BWCs than among citizens who interactedwith officers who did not wear BWCs.2. Individuals with voluntary officer interactions viewed officers as having greaterlegitimacy, professionalism and satisfaction than individuals who had involuntaryofficer interactions (see Figure 14), as expected.3. PERF researchers found no significant differences between citizens' perceptions ofofficers depending on whether the officer was wearing a BWC (see Figure 13). Thiswas contrary to the expectation that the presence of a BWC would result in betterperceptions of police legitimacy, satisfaction with their interactions, and views ofpolice professionalism.Overall, respondents had favorable impressions of their interactions with Arlington policeofficers (see Figure 13). With average perception scores typically greater than four out offive, the results are consistent with the Arlington community seeing the local police aslegitimate and professional. While BWCs did not improve these perceptions, that may bedue in part to the fact that there not a great deal of room for improvement. We cannot ruleout a potential positive effect of BWCs if this work were replicated in a jurisdiction withoutsuch strong baseline trust in the police.Policy ImplicationsAlthough BWC programs can offer many benefits, they are not a “cure-all” for problems inlaw enforcement. This work is consistent with a key finding in the previous BWC literature,namely the association between BWC usage and sharp declines in citizen complaintsagainst officers. While citizen complaints were significantly lower among the BWC-trainedofficers