Program Profile: Police Body-Worn Cameras (Turkey)

This intervention involves officers wearing cameras on their uniforms to increase citizens’ intentions to comply with police and improve their perceptions of police during traffic stops. The program is rated Promising. Drivers who encountered officers with cameras had statistically significant increases in both compliance and cooperation with officers and improved perceptions of procedural justice and police legitimacy, compared with drivers who encountered officers not wearing cameras. 

A Promising rating implies that implementing the program may result in the intended outcome(s).

Program Goals

Traffic officers are among the most visible members of the Turkish National Police (Aytac 2005; Kazu 2003), yet survey evidence suggests that Turkish residents hold more negative attitudes toward traffic officers and have diminished trust in them relative to nontraffic officers (Adaman, Carkoglu, and Şenatalar 2005; Aytac 2005). These negative attitudes and lack of trust in traffic officers could lead to citizen noncompliance during encounters, which may result in officer injuries or increased likelihood of police use of force (IACP 2004; Mastrofski, Reisig, and McCluskey 2002). In response to these concerns, a pilot program was launched in 2012 to outfit a limited number of Turkish National Police traffic officers with body-worn cameras to capture audio and video recordings of encounters between police and citizens, which was followed by a camera mandate for all traffic officers in 2014 (Adana Police Department N.d.). The goal of police body-worn cameras for traffic officers is to improve citizens’ perceptions of procedural justice and police legitimacy and increase their cooperation with police.

Program Activities/Key Personnel   

The Eskisehir Police Department in the Eskisehir province of Turkey consists of roughly 3,000 police officers who began using body-worn cameras in 2014, after the 2012 pilot program. In 2014, the department’s Regional Traffic Enforcement Unit consisted of 84 traffic police officers who patrolled in 16 officially marked vehicles and worked “8/24” shifts (8 hours on duty followed by 24 hours off duty). For each 8-hour shift, three teams consisting of two officers each were responsible for carrying out traffic enforcement at certain traffic checkpoints on Eskisehir Province highways. The officers marked highway checkpoints with cones and directed drivers where to stop. In the officer teams, one officer pulls over and stops vehicles, asks for documents such as driver’s license, insurance, or registration card and inspects cars, while the other officer stays inside the patrol car to record the traffic stop in the system and issue a ticket if necessary.

Regional Traffic Enforcement Unit officers who were assigned to wear body-worn cameras were instructed by their supervisors to activate the camera at the outset of traffic stops, to keep the camera running until the traffic stops ended, and to point out the body-worn camera to the drivers and notify them that the encounter was being recorded.

Program Theory

The concept of police body-worn cameras generating positive impacts on citizen compliance and cooperation with the police is supported by a few theories: specifically, deterrence and public self-awareness. First, when citizens know their behaviors are being recorded, they may be deterred from engaging in undesirable behaviors and thus may be more likely to follow established norms that dictate compliance with police directives and cooperation with police requests for assistance (Ariel 2016; Ariel, Farrar, and Sutherland 2015).

Self-awareness perspectives theorize that body-worn camera officers will be fairer, more respectful, and courteous to citizens during encounters because of the heightened self-consciousness of their behaviors; essentially, when people are being observed, they are more likely to follow social standards and comply with rules. The use of body-worn cameras could stimulate procedurally just behaviors by officers wearing the equipment, which could enhance citizen perceptions of police legitimacy and indirectly facilitate cooperation and compliance with the police during encounters (Demir et al. 2020; McCluskey et al. 2019; Reisig, Bratton, and Gertz 2007).

These potential behavioral changes in officers equipped with cameras are consistent with key elements of police legitimacy: public voice, neutrality in decisionmaking, trustworthiness, and treatment with dignity and respect (Ariel 2016; Tyler 2006).

Study 1

Demir, Braga, and Apel (2020) conducted a quasi-randomized controlled experiment to examine the effect of police body-worn cameras on citizens’ willingness to comply and cooperate with specific police instructions and their perceptions of procedural justice and police legitimacy immediately following their traffic-stop encounters. The study took place in the Eskisehir province in Turkey with the department’s Regional Traffic Enforcement Unit officers over a 22-day period from November 20 to December 11, 2014.

All traffic control operations were conducted in 2-hour blocks either between 9:30 a.m. and 11:30 a.m., or between 12:30 p.m. and 2:30 p.m., in two checkpoints that were randomly selected from the eight total checkpoints. Initially, two out of three teams of officers were first randomly selected and then randomly assigned to deliver the treatment condition (wearing the cameras during the 2-hour speed control operation) and control condition (not wearing the cameras during the 2-hour speed control operation). One of the randomly selected two traffic checkpoints and one of the time blocks also were selected using random assignment to determine which condition would be delivered. Subsequently, drivers were selected to treatment and control conditions based on the rotational assignment of traffic checkpoints and time blocks. Further, the Regional Traffic Enforcement Unit deployment policy required the rotation of traffic officers across the eight traffic checkpoints; therefore, the composition of the specific officer teams changed over the course of study period (31 distinct officers in 23 different configurations of the study teams delivered the treatment and control conditions).

Officers in both the treatment and control conditions conducted routine, business-as-usual traffic stops. The treatment officers were visibly equipped with body-worn cameras and were required to show the camera to the driver and explicitly notify them that: “The encounter today will be recorded through a body-worn camera.” After the traffic stop was completed, the drivers were asked to participate in a survey that measured their perceptions of the immediate encounter with the police and their more general perceptions of the police. The surveys were completed at least 50 feet away from the traffic stop encounter by undergraduate student surveyors who were neither police officers nor civilian members of the police department. A single item that captured the subject’s level of agreement with the statement, “I did as I was told by the police officer,” was used to measure specific compliance (citizens’ acquiescence to instructions during direct interactions with the police) during the traffic encounter. A latent variable was used to measure general compliance (refers to citizens’ overall obedience to existing laws and civil ordinances) and was based on subject agreement with two statements: “I would obey traffic rules”; and “This interaction would have a positive impact on my future compliance with the traffic rules.” A latent variable was then used to measure subjects’ cooperation vis-à-vis their level of agreement with these two statements: “I am willing to assist police if asked”; and “I am willing to work with police to try to solve problems in my community.” Increased collective agreement with the statements regarding the general compliance and cooperation variables were indicated by higher values on these measures. Seven items made up the procedural justice latent variable: 1) “Overall, the police officer was polite and treated me with respect during the interaction”; 2) “The police officer was fair when making the decision to stop me”; 3) “Overall, what the police officer did was based on the rules”; 4) “I felt the police officer would do the same to anyone in my situation irrespective of his/her status”; 5) “The police officer gave me opportunity to express my views during the interaction”; 6) “The police officer listened to me during the interaction”; and 7) “I believe that what the police did is for my own safety.” Three statements were used to examine traffic police legitimacy and general police legitimacy. The items used the same words but included “traffic police” specifically or simply “police” more generally: 1) “I have respect for (traffic) police officers”; 2) “I have confidence in (traffic) police officers”; and 3) “I trust (traffic) police officers.” Respondents indicated agreement on a five-point scale from “strongly disagree” to “strongly agree” for all items.

Seventy-one percent of the drivers in the treatment condition (who were exposed to officers wearing body-worn cameras; n = 299) participated in the post stop survey, and 74 percent of the drivers in the control condition (who were exposed to officers not wearing body-worn cameras; n = 325) participated in the post stop survey.

The analysis employed 15 control variables to assess balance between the treatment and control conditions in the main effects models, and then served as general controls in the path regression analyses. These covariates consisted of demographic predictors (gender, marital status, age), socioeconomic variables (employment status, education level, income), geographic variables (province of birth, urbanicity of residence), indicators of prior experiences with law enforcement (police acquaintance, prior police contact, prior traffic stop, prior traffic ticket), and characteristics of the stop itself (ticketed, time, location). There were no statistically significant differences on any of the control variables between drivers in the treatment condition and drivers in the control condition. Demographic characteristics of the traffic officers were not provided.

Linear structural equation models that adjusted for the control variables were used to examine the difference between drivers assigned to the body-worn-camera condition during their traffic stop and drivers assigned to the non-body-worn-camera condition, on general and specific compliance, cooperation, procedural justice, traffic police legitimacy, and general police legitimacy. The models employed cluster-robust variance estimation, where clusters were defined as officer teams. No subgroup analyses were conducted.

These sources were used in the development of the program profile:

These sources were used in the development of the program profile: