Build Safer Communities with Police Open Data (US0088)

• Lessons from Reformers: Open Police Data (December 1, 2020)

In response to recommendations of the President’s Task Force on 21st Century Policing, the United States is fostering a nationwide community of practices to highlight and connect local open data innovations in law enforcement agencies to enhance community trust and build a new culture of proactive transparency in policing. The Office of Science and Technology Policy and the Domestic Policy Council have been working on the Police Data Initiative in collaboration with Federal, state, and local governments and civil society to proactively release policing data, including incident-level data disaggregated by protected group. This work aims to improve trust, bring better insight and analysis to policing efforts, and ultimately co-create solutions to enhance public safety and reduce bias and unnecessary use of force in policing. Currently, 26 participating jurisdictions including New Orleans, Knoxville, and Newport News, are working side-by-side with top technologists, researchers, data scientists, and design experts to identify and overcome existing barriers to police efficacy and community safety. The Office of Science and Technology Policy and the Domestic Policy Council will continue to expand the Police Data Initiative to include additional jurisdictions. They will explore opportunities to work more closely with state partners and work to build out more resources such as playbooks and technology tools to help jurisdictions easily extract and publish data.

For details of these commitments, see the report: https://www.opengovpartnership.org/documents/united-states-mid-term-report-2015-2017/

✪ Commitment 36. Police Open Data

Commitment Text:

Build Safer and Stronger Communities with Police Open Data

In response to recommendations of the President’s Task Force on 21st Century Policing, the United States is fostering a nationwide community of practices to highlight and connect local open data innovations in law enforcement agencies to enhance community trust and build a new culture of proactive transparency in policing. The Office of Science and Technology Policy and the Domestic Policy Council have been working on the Police Data Initiative in collaboration with Federal, state, and local governments and civil society to proactively release policing data, including incident-level data disaggregated by protected group. This work aims to improve trust, bring better insight and analysis to policing efforts, and ultimately co-create solutions to enhance public safety and reduce bias and unnecessary use of force in policing. Currently, 26 participating jurisdictions including New Orleans, Knoxville, and Newport News, are working side-by-side with top technologists, researchers, data scientists, and design experts to identify and overcome existing barriers to police efficacy and community safety. The Office of Science and Technology Policy and the Domestic Policy Council will continue to expand the Police Data Initiative to include additional jurisdictions. They will explore opportunities to work more closely with state partners and work to build out more resources such as playbooks and technology tools to help jurisdictions easily extract and publish data.

Responsible Institutions: Domestic Policy Council (DPC), Office of Management and Budget (OMB), Office of Science and Technology Policy (OSTP)

Supporting Institutions: Law enforcement leadership from states, counties and cities, academia, foundations, nonprofit organizations and technologists

Start Date: Not Specified ....... End Date: Not Specified

Editorial Note: This commitment is a starred commitment because it is measurable, clearly relevant to OGP values as written, has transformative potential impact, and is substantially or completely implemented.

Commitment Aim

As a response to declining public trust in the police [518] and calls for greater transparency of policing activities, [519] this commitment aimed for the Office of Science and Technology Policy and the Domestic Policy Council to expand the Police Data Initiative [520]—a community of law enforcement agencies, technologists, and researchers who publish data sets on policing activities—to include additional jurisdictions, explore possibilities to collaborate more closely with state partners, and develop resources to help jurisdictions extract and publish data. In doing so, the commitment aimed to improve trust in the police, leverage data to improve policing efforts, enhance public safety, and reduce bias and the unnecessary use of force.

Status

Midterm: Substantial

At the midterm, the government had made substantial progress on this commitment. By June 2016 (slightly more than one year after the Police Data Initiative was launched), the initiative had expanded to include 57 total participating jurisdictions, [521] covering approximately 40 million people, [522] and had published 136 datasets [523] relevant for policing. In addition, the Police Data Initiative Leadership Team hosted 180 people from law enforcement and civil society to discuss lessons learned in the context of police data disclosure efforts. [524] Concrete evidence of efforts to develop resources to facilitate the extraction and publication of police data remained outstanding at the midterm.

End of term: Substantial

By the end of term, progress on this commitment remained substantial. While the initiative expanded to include additional jurisdictions and datasets, limited progress was made on developing data extraction and publication resources. More specifically, by September 2017, the Police Data Initiative had expanded to include 135 total participating jurisdictions, and had released 295 datasets, representing more than a 100 percent increase on both accounts. [525] The initiative has also begun linking to various technological resources to facilitate extraction, publication, and analysis of police data via the Resources section of its website. [526] One such tool, the Open Refine Data Cleaning Tool, can be used to explore and clean data, and link that data with various web services. However, the majority of resources profiled on the site pertain to data analysis, as opposed to the extraction or publication of data, and therefore are less relevant in the context of this commitment given its emphasis on the latter.

Did It Open Government?

Access to Information: Major

Civic Participation: Marginal

This commitment significantly opened government with respect to access to information, and opened government more marginally with respect to civic participation.

As described above, the Police Data Initiative’s release of 295 policing datasets across 135 jurisdictions represents an unprecedented effort to collect and publish police data in a centralized repository relative to the status quo that prevailed at the start of the reporting period (which roughly coincides with the Police Data Initiative’s launch). The publication of several data sets on the use of police force and officer-involved shootings is particularly noteworthy. The availability of this type of information is very limited in the United States, [527] despite public demands for greater disclosures. [528] In response to a request for comment on this commitment, a representative for the Police Data Initiative highlighted the importance of these datasets as a “starting point” for discussions surrounding national policing, with the goal of bringing transparency and data to those discussions. [529]

The quality of the data to which the public now has access nevertheless remains a prominent concern, particularly with respect to ensuring consistency in policing data terminology and the temporal and substantive coverage of policing data across different jurisdictions. With respect to the former, Seth Stoughton—a professor at the University of South Carolina Law School who conducts research on policing—noted during a National Public Radio (NPR) interview that touches on the Police Data Initiative that “police are all over the map with their stats. For instance, some of them count drawing their guns as a use of force while others don't. And if they're reporting whatever they consider to be a use of force, it can make it all but impossible to actually get a meaningful comparison” across policing datasets. [530]

With respect to data coverage, Jeff Asher, a crime analysis and journalist for FiveThirtyEight, highlights a related set of issues in his analysis of the availability of data on crime incidents reported by four Police Data Initiative cities (Charlotte, North Carolina; Indianapolis, Indiana; Newark, New Jersey; and Orlando, Florida), noting that some “have not yet begun placing current incident data online, are providing only historical data, or are posting information on only certain types of incidents,” impeding stakeholders’ ability to carry out more rigorous analyses. [531]

The Police Data Initiative’s public release of policing data also raises privacy concerns to the extent that individuals’ identities can be learned from the data. As described by Chief Technologist of the Federal Trade Commission Lorrie Cranor, during a panel discussion organized as part of a White House event Opportunities & Challenges: Open Police Data and Ensuring the Safety and Security of Victims of Intimate Partner Violence and Sexual Assault, “of particular concern is the possibility that people who access open police data may be able to identify crime victims or reveal their locations. For victims of domestic violence and sexual assault, this could put their safety and security at risk,” highlighting the need for the Police Data Initiative to develop guidelines that participating jurisdictions can use to “de-identify” data. [532] Together, these issues, which the Police Data Initiative has yet to develop plans to explicitly address, [533] mitigate against a coding of “outstanding” for this aspect of the commitment.

With regard to civic participation, the progress report highlights several instances in which the Police Data Initiative has begun to collaborate with the public on related initiatives, such as a youth-oriented coding event in New Orleans intended to engage youth in app-building efforts using Police Data Initiative data, [534] and an event organized by the Orlando Police Department and attended by representatives of the Police Data Initiatives and sexual assault and domestic violence victim advocates to begin discussing which “data can and should be made public.” [535] These engagements, while promising, are one-off activities. While the frequency of these activities is likely to expand over time, a more institutionalized engagement mechanism would be required to facilitate a further opening of government with respect to civic participation.

Carried Forward?

At the time of writing, the US government had not published its fourth national action plan, so it is unclear if this commitment will be carried forward. The government should nevertheless continue efforts to expand the number of jurisdictions participating in the Police Data Initiative and the public release of policing datasets, while making more concerted efforts to improve data quality and address relevant privacy concerns. The Police Data Initiative should also work to develop tools that facilitate the extraction and publication of data, complementary to the development of tools that facilitate data analysis.

[518] Jones, M. Jeffrey. “In US, Confidence in Police Lowest in 22 Years.” Gallup. See http://www.gallup.com/poll/183704/confidence-police-lowest-years.aspx. Consulted 27 June 2017.

[519] McKesson, DeRay. “Opinion: Washington Needs to Tell the Truth about Police Violence.” Washington Post. 16 June 2015, http://wapo.st/2vjn3xV.

[520] Police Data Initiative. “Homepage.” https://www.policedatainitiative.org/. Consulted 9 October 2017.

[521] See an archived version of the Police Data Initiative website from June 2016, available here: https://web.archive.org/web/20160617230746/http:/publicsafetydataportal.org/participating-agencies/.

[522] Wardell, Clarence and Denise Ross. “The Police Data Initiative Year of Progress: How We’re Building on the President’s Call to Leverage Open Data to Increase Trust between Police and Citizens.” Medium. https://bit.ly/2HpiJiZ. Consulted 27 June 2017.

[523] See an archived version of the Police Data Initiative website from June 2016, available here: https://web.archive.org/web/20160617230746/http:/publicsafetydataportal.org/participating-agencies/.

[524] Doom, Alyssa and Damian Ortellado. “Lessons learned from a year of opening police data,” Sunlight Foundation. 4 May 2016. https://sunlightfoundation.com/2016/05/04/lessons-learned-from-a-year-of-opening-police-data/Consulted 28 June 2017.

[525] Police Data Initiative. “Datasets.” https://www.policedatainitiative.org/datasets/. Consulted 24 September 2017.

[526] Police Data Initiative. “Resources.” https://www.policedatainitiative.org/resources/. Consulted 24 September 2017. See specifically the “Tools” sub-section.

[527] Caplan, Robyn et al. “Open Data, the Criminal Justice System, and the Police Data Initiative.” Datacivilrights.org. 27 October 2015, http://www.datacivilrights.org/pubs/2015-1027/Open_Data_Police_Data_Initiative.pdf.

[528] Grothaus, Michael. “The US Doesn’t Track Deaths By Police, So Citizens Are Doing It.” Fast Company. 18 June 2015, http://www.fastcompany.com/3045724/fatal-encounters-crowdsourcing-deaths-by-police.

[529] Written comments provided to the IRM researcher, 27 October 2017.

[530] Kaste, Martin. “Data Initiative Aims To Help With Police Force Transparency.” National Public Radio. 28 April 2016. http://www.npr.org/2016/04/28/475985461/data-initiative-aims-to-help-with-police-force-transparency. Consulted 30 September 2017.

[531] Asher, Jeff. “Which Cities Share the Most Crime Data.” FiveThirtyEight. 28 December 2015. https://fivethirtyeight.com/features/which-cities-share-the-most-crime-data/. Consulted 30 September 2017.

[532] See discussion in Cranor, Lorrie. “Open Police Data Re-identification Risks.” DigitalGov. 17 May 2016. https://www.digitalgov.gov/2016/05/17/open-police-data-re-identification-risks/. Consulted 30 September 2017.

[533] With respect to privacy concerns, Technology First, an initiative managed by the Safety Net Project at the National Network to End Domestic Violence whose mission is “exploring technology in the context of intimate partner violence, sexual assault, and violence against women,” released a set of resources on 15 June 2017 to help address privacy concerns in the context of open police data; the resources include background materials, guidance on incorporating privacy concerns into police data initiatives, and a webinar supplement on related issues. Articles announcing the resources reference the PDI, but the development of these resources does not appear to be a PDI initiative. See Technology Safety. 15 June 2017. “*UPDATED: Preserving Victim Privacy While Increasing Law Enforcement Transparency: Finding the balance with Open Police Data Initiatives.” https://www.techsafety.org/blog/2017/6/15/updated-preserving-victim-privacy-while-increasing-law-enforcement-transparency-finding-the-balance-with-open-police-data-initiatives. Consulted 30 September 2017. See also Technology First. “The Police Data Initiative, Open Data, and Victim Privacy.” https://www.techsafety.org/police-data-initiative-victim-privacy. Consulted 30 September 2017.

[534] Ross, Denice. “Does Open Data Build Trust? A Story of Demond, Police Data, and His Grandmother’s Recycling Bin.” Medium. 22 July 2015. https://medium.com/@ObamaWhiteHouse/does-open-data-build-trust-49ee4d400ba. Consulted 30 September 2017.

[535] City of Orlando. “City of Orlando Organizes Domestic Violence and Sexual Assault Data Preview Event.” 28 January 2016. http://www.cityoforlando.net/news/2016/01/city-of-orlando-organizes-domestic-violence-and-sexual-assault-data-preview-event/. Consulted 30 September 2017.