Fatal police violence by race and state in the USA, 1980–2019 | Hacker News Reader

Fatal police violence by race and state in the USA, 1980–2019

thelancet.com

19 pointsFnoord6 comments

Synthesise more than half the data, then decry the injustice it shows. This is not science. This is advocacy of a belief set.

They didn't synthesize any data.

> We compared data from the USA National Vital Statistics System (NVSS) to three non-governmental, open-source databases on police violence: Fatal Encounters, Mapping Police Violence, and The Counted.

Emphasis mine:

> We found that more than half of all deaths due to police violence that we estimated in the USA from 1980 to 2018 were unreported in the National Vital Statistics System.

Same as astrophysicists interpolating samples of a wide spectrum from a few spikes and declaring they're "searched the spectrum and found no signal".

With almost 20,000 police departments in the US, no national law, and inconsistent state laws regarding the recording of police violence there is always going to be data quality issues. There are dedicated people combing through local news reports around the US trying to document police violence, but it's a nearly impossible task without a strong national law requiring police departments to inform the justice department every time they discharge a weapon.

If you decry using estimated police death information and believe the problem isn't this bad you should join anti-police activists in demanding a national database of police violence. If you believe the numbers will show police violence isn't a problem then call for action to collect the information properly.

Agreed the data sucks (and is going to because the reporting is important to those reporting it); and efforts to collect better data are important and should be pursued. Estimation is a vital tool and will be even when there's better data.

I'm even willing to say I suspect these folks are right about the bias in police killings. I also suspect it has more to do with perceptions of economic status than skin color.

However, stretching statistics to cover opinions like this does their cause no favors, and cheapens their integrity whilst damaging their credibility.

While this piece is garbage, I'd really like to see this data broken down by county.

My hunch, and I could be wrong, so data would be nice, is that the problem lies almost entirely with urban police, which are (effectively) under the command of the city mayor.

Reply on news.ycombinator.com