Researchers Release Full Data from 2012 Police Body Cam Study

​In the debate over police body cameras, there is basically one study that proponents repeatedly reference: a 2012 experiment in Rialto, California, carried out by the local police chief and two academics from the University of Cambridge.

The topline results—that cameras reduced the use of force by 60 percent and civilian complaints against police by 88 percent—were released by the Rialto police and have been cited widely since then. However, the full details of the study weren't available until now.

In fact, the full study wasn't even published until November, where it appeared in the Journal of Quantitative Criminology.

The researchers weren't planning to release the detailed results publicly, but they changed their minds when "it became highly topical," says Fred Lewesy, a communications officer at Cambridge University, which funded the study through its Jerry Lee Centre of Experimental Criminology. He sent the text of the study to journalists last week.

"Highly topical" refers not just to Ferguson, Missouri, where President Barack Obama has asked for federal funding to equ​ip cops with cameras. It's also a response to critics who took the Eric Garner case—in which a police officer walked free after choking a man to death, even though the entire incident was captured on video—as evidence that body cameras don't work.

The comparison is invalid, the researchers said, because a spontaneous cell phone video recorded by a witness does not have the same effect as long term, institutionalized surveillance. (Something Motherboard's Jason Koebler also pointed out at the time.)

"The 'preventative treatment' of body-worn-video is the combination of the camera plus both the warning and cognition of the fact that the encounter is being filmed," Barak Ariel, one of the authors of the study, said in a release. "In the tragic case of Eric Garner, police weren't aware of the camera and didn't have to tell the suspect that he, and therefore they, were being filmed."

This argument also gets at one of the interesting details in the study that has been overlooked: Suspects were also aware of the cameras. This fact has been widely misunderstood. Margaret Talbot wr​ote in The New Yorker that, "it seems more likely that the police would be conscious of the camera's presence… than most civilians, who may not notice the green recording light."

The full methodology reveals, however, that officers wearing body cameras verbally warned suspects the encounter was being recorded at the start of every interaction—just like the start of a call with your credit card company.

That's just one of the interesting tidbits you'll learn if you tuck into the full text of the Rialto study.

Another finding that has not received much attention is the so-called "spillover" effect in the Rialto study. The researchers felt that, with only 54 frontline officers, the sample size would be too small if they just gave half the officers body cameras and compared the two groups. Furthermore, police officers work in pairs, and it would have caused scheduling complications to ensure that both officers either had cameras or didn't.

Instead, the researchers decided to have the same officers wear cameras half the time, dividing the officers' shifts into "treatment" (body cameras) and "control" (no camera) conditions. This gave them a much larger sample size—they ended up with 988 shifts—but it also created a spillover effect. The researchers actually found a significant reduction in use of force over both conditions. Officers used less force when they were wearing body cameras during the experiment than when they weren't. But even when they weren't wearing the cameras, officers were using less force than they had in years before the experiment.

"The reduction in use-of-force, coupled with a reduction in citizens' complaints, was registered across both study arms, which suggests that the effect of being observed during experimental shifts diffused to control shifts," the researchers write.

It's also interesting to note that the officers were regularly downloading and viewing the footage from their own body cameras, which may have had an additional psychological effect on top of being cognizant of the camera in the moment.

(Also fun to note: the cameras and video management system, evidence.com, were both provided by Taser. The company that makes the ubiquitous electroshock weapon now wants to make the ubiquitous cop body ​camera.)

Even though the Rialto study is the most comprehensive comparative experiment on police body cameras in the US to be released to date, the researchers conclude by saying more testing is needed.

"This is but one experiment and before this policy is considered more widely, police forces, governments, and researchers should invest further time and effort in replicating these findings," they wrote.

The researchers also caution that their experiment had the blessing of the police chief, and that body cameras may not work as well if the administrator of the program is "less influential in the organization."

When you look at the actual, hard number results of the Rialto study, the data feels scarce. During the experiment, there were a total of 25 incidents of police use-of-force. Seventeen occurred during control shifts—no cameras—and eight during experimental shifts. That's it. That's the source of the 60 percent reduction. Seventeen compared to eight.

What about other experiments? Two more studies were carried out in England and Scotland, but neither used a comparative design. In the US, an experiment by police in Mesa, Arizona, concluded two months after the Rialto experiment and will likely be published soon. In that study, also conducted over a 12-month period, 50 officers were assigned to wear body cameras and 50 were assigned to a control group; the initial findings were similar to the Rialto experiment. That study is under review and expected to be published early in 2015, the researchers told Motherboard.

Another experiment in Phoenix, Arizona, concluded in April of 2014; preliminary results suggest that the cameras reduced civilian complaints and led to more arrests, but details have not been released.

Even though so far only one rigorous study has been published, police body cameras are basically here. At least 63 agencies have been using body cameras in some capacity, according to a 2014 report commissioned by the Department of Justice, and police departments in Las Vegas, Spokane, and other cities are running significant pilot programs. The NYPD just initiated its own beta test. The President has thrown his weight behind the idea, as have federal judges.

Advocates may feel that the issue of police brutality is too urgent to wait for more data. The danger, however, is that these programs will be difficult to design effectively without a deeper understanding of why they work. (When police in Los Angeles were required to install audio recording equipment, they just ripped the antennas off.) Furthermore, cameras could have undesirable long term consequences—such as eliminating the ability to bring a case to court if there isn't video evidence.

"Body-worn-video has the potential to improve police legitimacy and enhance democracy, not least by calming situations on the front line of policing to prevent the pain and damage caused by unnecessary escalations of volatile situations," Ariel, one of the Rialto authors, said in a statement. "But there are substantial effects of body-worn-video that can potentially offset the benefits, which future research needs to explore."