None of their colleagues may have noticed, but a computer has. By churning through the police’s own staff records, it has caught signs that an officer is at high risk of initiating an “adverse event” – racial profiling or, worse, an unwarranted shooting.
The Charlotte-Mecklenburg Police Department in North Carolina is piloting the system in an attempt to tackle thein the US in the past three years. A team at the University of Chicago is helping them feed their data into a system that learns to spot risk factors for unprofessional conduct. The department can then step in before risk transforms into actual harm.
The idea is to prevent incidents in which officers who are stressed behave aggressively, for example, such as one in Texas where an officer pulled his gun on children at a pool party after responding to two suicide calls earlier that shift. Ideally, early warning systems would be able to identify individuals who had recently been deployed on tough assignments, and divert them from other sensitive calls.
Police departments have had early intervention systems for some time, but in many cases, they simply amount to assigning people specifically to watch out for officers who appear to be.
The system being tested in Charlotte is designed to include all of the records a department holds on an individual – from details of previous misconduct and gun use to their deployment history, such as how many suicide or domestic violence calls they have responded to. It retrospectively caught 48 out of 83 adverse incidents between 2005 and now – 12 per cent more than Charlotte-Mecklenberg’s existing early intervention system.
More importantly, the false positive rate – the fraction of officers flagged as being under stress who do not go on to act aggressively – was 32 per cent lower than the existing system’s. “Right now the systems that claim to do this end up flagging the majority of officers,” says Rayid Ghani, who leads the Chicago team. “You can’t really intervene then.”
“The most appropriate intervention to prevent misconduct by an officer could be a training course, a discussion with a manager or changing their beat for a week”
Ghani and colleagues will present theat the International Conference on Knowledge Discovery and Data Mining in San Francisco this month. They are now testing the system with the Los Angeles County Sheriff’s Department and Knoxville Police in Tennessee.
The big question is what to do once an individual has been flagged as a potential risk. The team is in the early stages of working with Charlotte-Mecklenburg police to find the most appropriate intervention, which could be a training course or a discussion with a manager. Or it might be more subtle, such as changing the officer’s beat for a week.
One thing Ghani is certain of is that the interventions will need to be decided on and delivered by humans. “I would not want any of those to be automated,” he says. “As long as there is a human in the middle starting a conversation with them, we’re reducing the chance for things to go wrong.”
Frank Pasquale, who studies theat the University of Maryland, is cautiously optimistic. “In many walks of life I think this algorithmic ranking of workers has gone too far – it troubles me,” he says. “But in the context of the police, I think it could work.”
Pasquale says that while such a system for tackling police misconduct is new, it’s likely that older systems created the problem in the first place. “The people behind this are going to say it’s all new,” he says. “But it could be seen as an effort to correct an earlier algorithmic failure. A lot of people say that the reason you have so much contact between minorities and police is because the CompStat system was rewarding officers who got the most arrests.”
CompStat, short for Computer Statistics, is a police management and accountability system that was used to implement the “broken windows” theory of policing, which proposes that coming down hard on minor infractions like public drinking and vandalism helps to create an atmosphere of law and order, bringing serious crime down in its wake. Many police researchers have suggested that the approach has.
Pasquale warns that the University of Chicago system, like any other, will probably include some biased data. For example, a black police officer in a white community is likely to get more complaints than a white colleague, he says, because the police can be subject to racism, too. Giving officers some channel to seek redress will be important. “This can’t just be an automatic number cruncher.”
More on these topics: