Ask The Right Questions

People are asking a lot of questions about what police do, and whether they do things right.

Answering important questions about law enforcement being asked today requires a collaborative approach that starts before data is even gathered.

Part of the problem is police culture. We’re just not that good at data. We lack the tools and training to make sense of the mountains of data we collect. We also hate speaking when we’re uncertain. Until recently, no cop got in trouble for not saying something. That’s changed in the past 10 years with the rise of data journalism, but the instinct remains.

Cops usually are uncomfortable with change. Most police procedures are established to prevent repeating a tragic mistake. Things that alter those procedures cause a real fear of unknowable, unintended consequences. We hate that. To the rest of world, this looks like defensiveness, but really it’s a fear that doing something other than what proved to work in the past brings risk – risk of life and of limb.

Another challenge is that public expectations are high. Television police procedurals have groomed people to expect that police can easily get any data they need. That’s wildly unrealistic. Television programs misrepresent woefully the technological capabilities of police services.

Cases of deadly force are among the most important and the most rare. In the United States, about 0.02 percent of face-to-face encounters result in a use of deadly force. The other 99.98 percent of the time, police are just policing.

Yet, despite how infrequently deadly police encounters occur, they command tremendous public interest. Any time a life is lost the incident and relevant practices deserve special scrutiny.

The question people ask most often is, “How often do you shoot people, and what are their races and genders?” That’s a ham-fisted proxy for the real concern, “Are you treating our communities fairly?”

When one aspect of policing become so all-consuming as to drive out attention to others, oversight doesn’t work and data can’t keep up.

We should instead ask very straightforward questions:

Are my police treating my community fairly?

Are our police engaging in activity that we approve of?

Are they racially profiling?

Are they not racially profiling?

If we can get to the data that covers how we are coming into contact with people, and the outcomes of those contacts, we can start to answer these questions.

We must do it, though, with an open mind, not a set series of expectations. We must be data travelers, and see what we see in the data, as opposed to data tourists, who see what they came to see.

We must take fearless and careful and considered looks at all the data – not just the data that addresses a certain topic like race or gender. All of it. Once we look at the available data, then we can see how the data informs our policies.

What are the behaviors of and conditions experienced by officers that are predictive of violent encounters?

What are the behaviors of and conditions experienced by civilians that are predictive of violent encounters?

Only when that’s done, can we set about making policies informed by these insights. And we absolutely should come up with a list of things our officers must “Don’t Do”, but we must also come up with a list of things our officers must Do.

In other words, instead of making a long, and increasing, and never-ending list of things that a police officer should not do, we could actually start to look at the outcomes of people who are doing well and start to make shorter lists of things that they should be doing.

Best practices, informed by good outcomes. Best practices that are informed by the data. Not of what we think will happen, but what we know has happened – and not self-reported data, but data that they’re collecting just as part of doing their day-to-day work.