The use of force by police officers is currently a matter of great public interest. From 2018 to 2022, thanks to an award from the Office of Community Oriented Policing Services (COPS Office), the City of Rochester (New York), the Rochester Police Department (RPD) and the Rochester Institute of Technology (RIT) collaborated in order to study incidents of use of force by the RPD.
The main goal of the project was to help the public understand the RPD’s use of force that went beyond engaging with viral videos. To guarantee the privacy of both officers and others involved in the RPD’s use of force, as well as to produce a large number of incidents, the researchers aggregated the use of force incidents into one dataset.
The report, published by the U.S. Department of Justice, highlights that the creation and analysis of this data set also enabled the RPD and RIT to pursue the secondary goal of uncovering potential improvements in RPD policy and training on the use of police force.
The purpose of this paper is to help other agencies replicate this project with their own data on use of force. Therefore, the paper sets out not only the findings of this project, but also the processes used and the challenges the researchers found in the hope that other police agencies can use and extend them.
Thus, it is relatively easy to access data on each use of force incident for the purpose of examining a single incident, but extremely difficult to aggregate them into a data set of all use of force files.
Due to the inaccessibility of the data, a significant part of that project was the transformation of a PDF image folder into a row and column dataset.
After creating the dataset, the researchers undertook two tasks: first, describing that dataset for a lay audience and, second, gaining new insights into the RPD’s use of force.
To help other law enforcement agencies to replicate this analysis, the researchers explain several organizational and bureaucratic challenges they encountered that other agencies could potentially avoid or mitigate.
This report concludes with recommendations for different police agencies considering this or similar attempts to learn from existing local data analysis, which are as follows:
- Law enforcement agencies should establish relationships with universities in the area of research.
- Law enforcement should store data digitally whenever possible.
- Police analysts should include the citizenry as one of the many clients they serve.
- Law enforcement agencies should invest in computer resources capable of performing advanced analytical techniques.
- Local government agencies should consider the enormous potential of using the latest IT tools.