Analysis of Police Use of Force Data

I collected a lot of data. We collected millions of observations

on, uh, everyday use of force that wasn't lethal.

We collected thousands of observations on lethal force.

And, and it, it was in this moment in 2016

that I realized

people lose their minds when they don't like the result.

And so what my paper showed, you'll see tomorrow, uh, s-

like some of you,

uh, was that, yes, we saw some bias in the low level uses of

force every

day pushing up against cars and things like that.

People seem to like that result, but we didn't find any,

um, uh, racial bias in police

shootings. Now, that was really surprising to me because I

expected to see it. The little known fact

is I had eight full-time

RAs that it took to do this

over nearly a year. When I found the surprising result,

I hired eight fresh ones and redid it

to make sure.

They came up with the same exact answer, and I thought it was robust, and then I

went to go give it, and my God, all hell broke loose.

It was a hundred and four-page dense

academic economics paper with a hundred and

fifty-page appendix. Okay?

It was posted for four minutes when I got my first email. "This

is full of shit.

Doesn't make any sense." And I wrote back, "How'd you read it that

fast?"

"That's amazing. You are a genius." And

I had colleagues take me into to the side and say,

"Don't publish this.

You'll ruin your career."

Mm.

I said, "What are you talking about?"

I said, "What's wrong with it?" "Do you believe

the first part?" "Yes."

"Do you believe the second part?"

"Well, it's... the issue is they just don't fit together.

We like the first one, but you should publish the no-

the second one another time."

I said, "Let me ask this.

If the second part about the police shootings..." This

is a literal conversation.

I said to them, "If the second part,

um,

showed bias, do you think I would-- should publish it then?"

And they say, "Yeah, then it would make sense."

And I said, "I guarantee you I'll publish it.

We'll see what happens."

So it was, it was, you know... I, I lived under,

under, um, police protection for about thirty

or forty days. I had a seven-day-old

daughter at the time. I remember going and shopping for...

'cause, you know, when you have a newborn, you think you have enough diapers, you

don't. So I, I was going to the grocery store to get

diapers with the armed guard. It was crazy.

It was really truly crazy.