And here’s a report from @HFGuggenheim on some traits to consider for predicting mass violence (tho I think it risks some of the selecting-on-the-DV concerns I noted yesterday): hfg.org/wp-content/upl…
This is by no means an exhaustive list.
But as viscerally appealing and cathartic as making “Triangle Shirtwaist Academy” jokes is (and it’s a good one), this is what we need to do now.
Don’t focus on the distracting unserious proposals. Force the debate where it needs to go.
Another study, this one by @RANDCorporation, surveying the literature on what reduces gun violence for several different types of such violence (crime, suicide, etc.).
School shootings, while devastating, are rare. Mass shootings are too (tho can depend on the def).
So if we change hiring practices for these events, we are also changing who does all the routine policing where these officers will spend 99.99% of their time.
So: some data that is highly relevant to the @chesaboudin recall debate.
Here's an overlay of local voting outcomes with shootings in 2019. What we see that, by and large, Boudin did best where the shootings ARE.
This is pattern we see elsewhere: it was the case in Philly last year, and preliminary results show similar patterns in Chicago and even Pittsburgh (where the progressive lost, but still appears to have done better where there were more shootings).
This is important.
This undermines a common pushback against progressive prosecution, that it is the indulgence of progressive white voters in safe areas who are unaffected by their decisions.
I know there is a sense that “nothing after Sandy Hook = nothing ever.” But that’s… an assumption. It MAY be true, but we have no data.
And here—we have video, and video arriving under a different underlying politics-of-crime.
Maybe nothing will come of this. But many Black men were murdered by police before George Floyd—yet his murder was ultimately a match that caught. (Of course, for some reason the others didn’t; it’s all so hard, if not impossible, to predict.)
Just need to say: graphs plotting some sort of gun control score vs gun deaths—all of which show more regs = less deaths—don’t… really show that. And may in fact misshape our debate on gun violence.
My priors remain that regs reduce violence. But….
There’s a huge selection effect problem here. The states that adopt tougher regs aren’t random. They’re the states more likely to invest in other social services as well.
Those TOO reduce gun violence. Maybe—given our volume of guns, perhaps likely—more than gun control.
Those simple graphs don’t show the impact of, say, Brady Scores on gun death. They show the impact of being in a state w a low Brady Score—including all the OTHER stuff that correlates w that score.
But “better Brady = less death” puts crim legal responses at the center.
We have never been in a political moment like this, and we basically have no studies on what the (actual) politics of crime are to try to extrapolate from in the first place.
It’s way way WAY too early to say if reform has died or not.
Going out on a limb to guess that most countries have not been forced to set up a database focused exclusively on gun violence in K-12 schools.
That we have--like the fact that we have to have methodological debates over what counts as a "mass shooting"--speaks ... volumes.
If interested, you can check it out here: chds.us/ssdb/
"How do we separate `school shootings' from all the other ways gun violence permeates our schools is a tough methodological question..." asks one of the only countries that has to have this methodological debate.