Killings by US Police Higher Than FBI Reported, New Federal Program Shows

In 2015, the US government began using a redesigned program to count killings by police, and it has revealed that the FBI has been vastly under-reporting the number of deaths.

The Justice Department’s Bureau of Justice Statistics (BJS) released the findings of a new study on December 15, and explained the enhanced data collection methods that are now being used:

…the redesigned ARD program has two phases to its data collection methodology to maximize completeness. In the first phase, BJS reviews open information sources, including news outlets and official agency documents, to identify potential arrest-related deaths. In the second phase, BJS surveys law enforcement agencies and medical examiner/coroners’ (ME/C) offices to confirm all arrest-related deaths occurring in their jurisdiction and collect additional information about those deaths.

The program, which draws on data collected by The Guardian, recorded 270 homicides by officers in three months last year. The FBI said earlier this year that it had counted just 442 in all of 2015.

The FBI’s annual count of homicides by police, which depends on departments voluntarily submitting their numbers, was discredited after it became clear that the method was recording less than half of all killings nationwide.

From The Guardian:

The BJS trial closely examined all deaths in custody – including suicides, accidents and some others – from June to August 2015. It initially identified 377 deaths through media reports and other publicly available sources, 268 of which were found to be homicides by officers.

Researchers then sent surveys to almost 500 police agencies and medical examiners linked to the deaths identified in their open-source review, asking them to confirm the deaths and inform them of any others. During this second phase, researchers identified an additional 48 deaths that were not reported in the media, two of which were homicides.

Duren Banks, the program’s lead statistician, told The Guardian that the trial had been “a success” compared to a previous incarnation, which was estimated to be catching only half of all arrest-related deaths before being temporarily shut down in 2014.

“We really went back to the drawing board to see what was working and what wasn’t, and to try to get a more comprehensive count of arrest-related deaths in the United States,” said Banks.

Accurate reporting of deaths involving police is crucial to understanding the depth of the problem so improvements in relations between citizens and police can be made.