IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Can Good Data Improve Policing and Prevent Excessive Force?

The police force of Durham, N.C., has teamed up with SAS Institute to create a data system that will put a spotlight on exemplary police work and reveal cases where officers may need training or counseling.

An officer standing in front of a police car with its lights on.
Police departments have faced increasing scrutiny over the last decade, with public criticism reaching a peak after a Minneapolis police officer brutally killed George Floyd while he was in custody last year. How should law enforcement agencies encourage better police work? Durham, N.C., intends to answer this question with a data system that will inform supervisors about officers who exemplify best community practices or display warning signs of inappropriate conduct. 

Durham Police Department (DPD) has used data and analytics for a long time and recently looked at the prospect of centralizing officer performance metrics with in-house resources, said Jason Schiess, DPD analytical services manager. The agency realized that SAS Institute, which has worked with DPD before, was better positioned than DPD to create officer performance dashboards because of its ability to bring “together data from different disparate sources into a central environment.”

The new system will offer a comprehensive view of officer performance. Traditionally, performance is measured by variables like answered service calls, citations written and arrests made, but these factors only represent “a very small component of the totality of an officer’s activity,” Schiess explained. 

The goal is to go beyond such factors and capture positive community interactions in order to promote “the things that the agency wants to champion and reinforce.” To illustrate what one of these interactions might be, Schiess cited the simple example of an officer playing a game of basketball with kids on the street. 

The system will also provide data on variables that may indicate that an officer is at risk of making bad decisions. Such factors can include use of force, complaints and use of personal leave. 

“We’re looking at performance and behavior indicators that, when combined, provide insights into an officer that might be in a category of risk,” said Juan Colon, a former police major who now serves as an advisory industry consultant for SAS. “Once the system identifies that, the supervisor receives an alert and is able to intervene with training and counseling to make sure the officer stays on the right path.”

Colon, who was in law enforcement for more than 28 years, said an officer will experience events that can trigger tremendous amounts of stress that can then lead to a tragic situation like one would see on the news. Colon recalled events that deeply affected his own well-being, like when he watched an 11-year-old die, but his department had no tool to understand what he was going through. 

“They had no idea how these incidents were affecting me personally,” Colon said. “I can tell you that those are situations that I think about frequently. It would have been great to see how my behavior may have changed after those incidents occurred. I see what has happened to the law enforcement community over the past few years. I’m a firm believer that a lot of these officers that are taking the inappropriate actions that they have been taking has been a direct result of their wellness not being in place.”

“Frontline supervisors don’t have the wherewithal to understand every one of their personnel,” he added. “You need technology to be able to do this.”

Schiess pointed out that police departments have a ton of data but don’t necessarily have a method to utilize it and improve performance. Thus, the most promising thing about the system is that it “empowers an agency to be proactive and not wait for an adverse event in order to act.”

DPD is in the beginning stages of deploying this data initiative. The plan is to implement the system this year in stages. Some of the data sources will involve more privacy issues, which can raise concerns among officers. 

Colon said SAS has met with police unions about the system. Some union members are hesitant about the idea until they see the positive benefits for the law enforcement community. The larger point for local departments is to understand that certain types of data can be quite sensitive. 

Another goal behind DPD’s data initiative is building public trust through improved transparency. Schiess said the system will have a public-facing component so that citizens can see how well officers are performing. DPD wants residents to have a fuller view of what their police force does. 

“They [the public] tend to fill the unknown with myths,” Colon said. “This is why transparency is so crucial.”


Jed Pressgrove has been a writer and editor for about 15 years. He received a bachelor’s degree in journalism and a master’s degree in sociology from Mississippi State University.