BPC-CPD Corpus
Description of Corpus Content:
The Broadcast Police Communications of the Chicago Police Department (BPC-CPD) corpus includes ~46 hours of manually transcribed two-way radio communications about CPD activity in 2018-19. It is composed of over 62,000 radio transmissions during this period. See IEEE SLT 2024 for further details.
Citation Information
Access to BPC-CPD Corpus:
Request access to the BPC-CPD corpus by emailing a completed version of this data use agreement (datasheet adapted from Agnew et al. 2024) to Chris Graziul. Following ethical review, your request will be granted or denied. If your request is granted then you will be provided a copy of the countersigned agreement and instructions for downloading BPC-CPD.
Be sure to consult with relevant institutional administrators to determine who has the authority to sign this agreement!
Part of this data use agreement includes a Project Description that must be provided for access to be granted. Proposed activities will be evaluated for ethical considerations. Access to BPC-CPD will not be provided for activities considered unethical or ethically questionable. Denial of access on these grounds will be communicated via correspondence regarding your request. Those denied access are invited to reach out to discuss if an acceptable revision of your request may be possible. The purpose of this review process is to ensure adherence to Belmont Report principles. It is not meant to prevent access to BPC-CPD nor to provide preferential access to BPC-CPD for specific research topics.
To facilitate successful requests for access, below are basic guidelines for generating your Project Description:
Guidelines for Generating a Project Description
Your Project Description is the most important part of your access request. It should explain (1) what you plan to do with the corpus, (2) how your planned research activities adhere to Belmont Report principles, and (3) why access to BPC-CPD is necessary to conduct these activities. Your request will be evaluated according to the completeness and quality of these explanations.
If generating this content is difficult then please seek assistance from your local Institutional Review Board (IRB) regarding the contents of an IRB study protocol. You are not being asked to submit a protocol for review, but you may need to obtain IRB approval from your institution as part of using BPC-CPD so it is desirable to engage IRB personnel about their expectations. Related, it is vital to understand IRB review as an important evaluation of proposed research activities whose ultimate purpose is to enable these activities. For clarity, your Project Description will not be formally reviewed by the University of Chicago’s Social & Behavioral Sciences IRB but they may be consulted as part of the review process.
Please reach out about your Project Description prior to submitting your access request if your proposed project does not fall within the following use cases: replication or model evaluation, educational use, model development (with no plans for model deployment). Access to BPC-CPD for other use cases will likely require additional review by University of Chicago’s University Research Administration. See the above data use agreement before submitting such a request, as some uses are explicitly prohibited.
DON’T PANIC! These Guidelines are designed to facilitate access requests. If they are unhelpful or any part of the access request process is unclear then you are strongly encouraged to reach out for clarification.
Annotation guidelines:
The final protocol used by annotators to generate the BPC-CPD corpus may be found here
Replicability:
Technical details to replicate findings will be provided in the future and are available on request.
IEEE SLT 2024
National Repository of Policing Data (NRPD)
University of Chicago, SBS IRB23-0660
The National Repository of Policing Data (NRPD) was established at the University of Chicago in 2023 to provide secure and responsible public access to policing data. Access to BPC-CPD is provided through NRPD in accordance with its reviewed IRB protocol and related processes for ensuring the protection of human subjects and adherence to Belmont Report principles.
Principal Investigator: Chris Graziul
Impact statement:
The BPC-CPD corpus is composed of data considered and treated as publicly available in the State of Illinois. This does not negate the ethical responsibility of researchers to ensure adherence to Belmont Report principles. Creation of the BPC-CPD corpus was supported by an award from the National Institute on Minority Health and Health Disparities of the National Institutes of Health (R01MD015064). Given this source of funding, related requirements for data sharing, and the sensitive nature of BPC-CPD content, access to BPC-CPD is provided to users following ethical review of research use. See Venkit et al. (2024) regarding known racial/ethnic and gender disparities in privacy vulnerability associated with BPC in the city of Chicago.
Updated February 3, 2025