Police Data Initiative
This project was a part of our Citizen Interaction Design class. We partnered with Ferndale, Michigan's police department to help them increase transparency and trust with their community through police data transparency. Our team helped the police department understand the information needs of their community, and developed an open data tool to showcase information deemed as relevant by the Ferndale community leaders and members.
In the light of the national climate in the United States regarding police brutality, the lack of transparency into police activity has impacted the trust between the police and the community. As part of their implementation of 21st Century Policing, the Ferndale Police department is participating in the Police Data Initiative, a program of the US digital services at the White House that supports over 100 police departments across the country in increasing transparency of police interactions, arrests, and other police data. Now the department needs to establish the policy, tools and processes to facilitate and maintain that data for the public while respecting privacy concerns and maintaining law enforcement standards.
Understanding the Data Landscape
We worked extremely closely with the Community Policing Sergeant to learn about the Ferndale police department's current data landscape, the tools they currently used to maintain the data, and processes they had in place to input and extract this data. We also looked into use cases and frequency of data requests (ex. FOIA requests). To gather this information, we interviewed the records and IT personnel at the department.
Open Data Tool
In order to understand things from the data portal perspective, we talked to companies such as Socrata, Code for America on best practices on which tools and processes to use to establish an open data portal, and finally the city's IT department to see if they had established partnerships with any of the open data organizations. We found that the city's IT department did have a partnership with ArcGIS open data, so we worked closely with CLEMIS, the tool that was used to store the police data, to see if and how they could potentially be integrated.
In our conversations with the police officers, we found that there was no ongoing engagement mechanism between the police and their community. We as a group wanted high community engagement throughout the entire police data initiative, since this data portal was being built for the community. To learn more about best practices around community engagement and approach, we contacted Northampton Police Department (PD), who had already gone through implementing an open data portal and an ongoing mechanism for community engagement. From the best practices we learned, we reached out to elected officials, community leaders, police officers, professors, and interested community members to form an open data task force to discuss the plan for this city's police data initiative.
Prior to the first meeting with the open data task force, we worked with the Community Policing Sergeant to extract a data set we believed the community would find useful. We settled on Prisoner and Police Demographic data because it was easy to extract, required little clean up, and was interesting. All information aligned to personally identifiable information was removed. We decided to show this data set, along with several others that were extracted from open data portals from other PDs, to illustrate examples of different types of police data sets and gauge the community's opinions on the usefulness of these data sets.
At the first meeting, we first had the community members introduce themselves and talk about why they wanted to be involved in this initiative to create a shared sense of purpose and goal. We then educated the members on what open data meant, and the importance and benefits of having an open data portal to share police data in the community. Next, we handed out samples of police data that were published across the country in addition to the prisoner data and police demographic data that we had extracted. Discussion from this led to feature dot voting to identify important data sets according to the task force, and prioritize which were more important to publish sooner.
Open Data Task Force members at the first police data initiative meeting
Feature dot voting. Community members walking around and putting sticky notes on data sets they deem important to publish
Through this project, Ferndale was only the second city in the country to create an ongoing community task force for accountability and collaboration around open data. The community meetings occurred every month with growing community members joining the open data task force to voice their opinions and lead the way for the city's police open data initiative.
We ended up publishing the dataset that we worked with and had access to in the period of 3 months. A team member continued to work for the city, and carried out our previous work and plans for the open data portal. The data portal is now in beta with Police Demographic and Crime Mapping data. You can view it at this website: http://data.ferndalemi.gov/