While researching police shootings, I interviewed (via email) Andrea Burch, the statistician who is currently in charge of the Bureau of Justice Statistics' Arrest-Related Deaths Program which purports to track the number of people killed by the police each year. Below, I have published her responses to my questions to provide some insight on how BJS gathers its data.
Interview with Andrea Burch
Q: When was the arrest-related deaths program started?
A: In response to the Deaths in Custody Reporting Act (DICRA) of 2000, BJS initiated the ARD program in 2003. The ARD program was the third of the three data collections that BJS developed in response to DICRA, the other two were collections of deaths in local jails and in state prisons. A more detailed history of the program can be found on the ARD website under the “About ARD” tab: https://www.bjsard.org/AboutARD.aspx
Q: What methods are used to gather the data? How are police departments made aware of the program? Is there a vetting process after the data is collected?
A: BJS collects the data through State Reporting Coordinators (SRCs) in each state and the District of Columbia. The SRCs are responsible for identifying and reporting to BJS information about each death in the process of an arrest. SRCs are responsible for identifying deaths that are within the scope of the ARD program, collecting information about these deaths, and transmitting this information to BJS through submission of the CJ-11 Arrest-Related Deaths Summary Report and CJ-11A Arrest-Related Death Incident Report.
SRCs use a variety of methods to identify ARD-reportable deaths, including relying upon centralized reporting systems on violent deaths that occur in their states, using direct contacts with law enforcement agencies, obtaining information from medical examiner/coroner’s offices, and/or conducting systematic open-source searches. Once SRCs identify an arrest-related death, they follow-up with the law enforcement agency involved with the death to obtain the data necessary to complete a CJ-11A Incident Report. In instances where the involved law enforcement agency does not complete the CJ-11A form, the SRC completes the form using information from official source documents (e.g., police report/press release, death certificate, legal proceedings).
Law enforcement agencies are made aware of the ARD program through contact with SRCs. Some SRCs will proactively contact their state’s law enforcement agencies and distribute information about the ARD program. In this instance, SRCs encourage law enforcement agencies to take a direct role in identifying deaths and reporting information. Other SRCs will take a more targeted approach and contact law enforcement agencies after an arrest-related death has been identified.
In addition, law enforcement agencies are made aware of the ARD program through communications from BJS and members of professional law enforcement associations. For example, ARD program staff (i.e., BJS, BJS’ national data collection agent) may reach out to a law enforcement agency and request participation in the program. In addition, BJS staff may discuss and distribute materials about the ARD program during meetings with law enforcement officials. Law enforcement agencies may also become aware of the program through information disseminated through professional associations. For example, there might be a presentation about ARD at an annual meeting or an article about the program in an association newsletter or other publication.
Data are submitted to BJS by each state’s SRC. The SRC offices are typically located within state criminal justice commissions (commonly administered by the governor’s office) as well as in state attorney general offices, state police departments, and state medical examiner offices. All data are collected on a standard CJ-11A Incident report and can be completed through the use of both paper and electronic forms. SRCs completing the CJ-11A Incident Report in place of a law enforcement agency may obtain information about the death through a phone interview and then record the information in a standard CJ-11A form. Submitted data are vetted by ARD program staff.
Q: Is the data a fairly accurate representation of how many arrest-related deaths occur or are there any problems with under-reporting or over-reporting?
A: BJS is undertaking a study to determine the scope of coverage of the ARD data and to determine if there is undercoverage.
Q: Which law enforcement agencies submit data? Does the program cover local police, state police, campus police, sheriff's departments, and federal law enforcement, or only some of these types of police agencies?
A: The ARD program covers all state (e.g., State Police, Highway Patrol) and local (e.g., police departments, sheriffs’ offices, campus police, transit authority). BJS does not release information about specific agencies reporting to the ARD program. Information about the number of agencies with recorded arrest-related deaths is presented at the state-level in the BJS publication, “Arrest-Related Deaths, 2003-2009 – Statistical Tables” (Table 11, pg. 10). Federal law enforcement agencies are not within the scope of the ARD program and deaths occurring during the process of arrest by federal officers are not recorded in the ARD collection.
Q: Are there any changes that will be made to the program in the future? Are there any changes you would like to make to the program, but lack the resources to make?
A: BJS is undertaking a review and assessment of the program. When the review is complete, I am charged with using the information from the assessment to make recommendations to the BJS director about the program.
- Christopher J. Mumola, "Arrest-Related Deaths in the United States, 2003-2005," Bureau of Justice Statistics, October 11, 2007 (mirror available here)
- Andrea M. Burch, "Data Table Arrest-Related Deaths, 2003-2009 - Statistical Tables," Bureau of Justice Statistics, November 17, 2011 (mirror available here)
- Arrest-Related Deaths Program Summary
- CJ-11A (form)