NoImmatureAdHom

Thanks for being curious!

You can follow the edweek links to look at their data. It gives enough detail on any given incident for you to make your own call about whether it's included for political reasons or actually representative of the type of thing parents might fear.

Mother Jones maintains a database of mass shootings int the U.S. that would include school mass shootings, and they (to their great credit) use non-crazy criteria to determine what to include: https://www.motherjones.com/politics/2012/12/mass-shootings-...

That's a cumulative 0.7% chance over 18 (or 22?) years, or 0.035%-ish per year. Not only is that not the right way to think about things even if shootings were random, the numbers aren't correct.

The Edweek numbers are doing what they were designed to do, which is mislead you and parent / grandparent. They are advocacy numbers, deliberately distorted to make things seem as bad as possible in order to drive clicks to further their political agenda. Which is exactly what's happening here!

You could read the edweek links, or the comment by jlawson, or you could do some math yourself. School shootings (in the meaningful sense, not the BS sense that edweek uses--shootings where a student is killed, which is what people really mean) are vanishingly rare in the U.S., a nation of 300,000,000 people. "Preparing" for them is the result of innumeracy, anxiety, fear-mongering, profitability of reporting on them, and lack of critical thinking skills in the U.S. population. The risk is so small that counteracting basically any other risk is a better use of resources. Teach kids about drugs. Teach them about driving safely. Teach them about how to be safe around pools, and how to swim. Teach them to eat healthy. Build them some exercise habits.

An hour invested in any of these activities is much better-spent than an hour of hand-wringing about an almost nonexistent risk of your kid being shot in school in the U.S. It's simple math.

I would add that many of the "school shootings" counted in the edweek.org data are not what most people would understand to be school shootings. Things like Person A shoots Person B outside of a school, nobody dies, neither A or B are students, teachers, or otherwise affiliated with the school. Edweek is providing what are known as "advocacy numbers", in that they are stretching the truth as much as they possibly can.

Your kid's chances of being killed in a school shooting are about the same as being struck and killed by lightning (many more people are struck and survive).

You know, you're right.

I think when you live it every day it gets exhausting.

This is a really stupid paper. Their effect sizes are tiny (a 3-4" deviation??). I don't care how accurately you estimated it.

Also, they don't talk about the things that are obviously driving this. It's not race per se, it's things that correlate with race. If you live in an environment where all black and non-black people have the same distributions of behavior, you won't learn to treat people differently on the basis of being black vs. non-black. It's the fact that black Americans are ~7-10x as violent[0] as non-black Americans that is driving the behavior the paper observes. Obviously. The fact that they don't touch this obvious driver makes them, in my view, cowardly, bad scientists out for a cheap thrill.

0: https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/explorer/crim... 7.5x raw numbers, closer to 10x when accounting for the fact that clearance rates are lower among black offenders (i.e., black Americans get away with murder more often).

The problems grandparent describes are real.

As a scientist, if you find "Black people good, white people bad" --> publication + promotion

You find "White people good, black people bad" --> you lose your job

Completely orthogonal to the morality of it, it creates something called publication bias (or the file drawer problem) in the social sciences. It's a huge issue.

Piracy is alive and well. I wonder why people pay for so many streaming services...

Yeah, I agree with this. Visiting SF from other American cities that have nothing like the scale of its problems...is jarring. I'm sure you can get used to it.

Also, we should differentiate between the feeling of comfort / safety (which is what most people care about), and actual crime statistics (which are themselves fraught, I don't think people in SF even report car break-ins these days since the police and justice systems don't work any more).

I don't know where you got your gun injuries / death stats, but there are two things to be careful of here:

1) substitution effects. As a hypothetical example, we could have one city that has 30 gun murders per 100k and another that has 130 per 100k, but both have the same murder rate (because you can murder with things other than guns). IMHO the important rate is the murder rate, and how much of it is with guns vs. other stuff is an aside.

2) Make sure your numbers don't include suicides. If they include suicides, they're not informative.

Their promises are worthless unless all client-side software is open-source and available for inspection with reproducible builds. It is unfortunate that WhatsApp managed to get significant network effects so fast (especially in Europe), meaning that many people feel the "have to" use it. If you "have to" use it:

0: encourage people to switch to Signal, Session, Briar, Matrix, etc.

1: you can use matrix bridges and similar things so you don't have to have Facebook's spyware on your phone. Examples:

-beeper.com (pulls together many messaging platforms, WhatsApp Twitter Slack IRC Instagram Telegram...) (can self-host)

-element.io operates another matrix-bridge-for-hire (can self-host)

-texts.com does something similar, but I don't know the details

-probably more I don't know about :-)

A very small percentage of the modern Western population. There are still populations on earth where basically every adult man hunts.

> If you have an actual sound ecosystem, in a sufficiently large habitat area (not encroached by humans, roads, etc.), culling is rarely necessary.

It's really a question of the extent to which humans get bothered when prey populations get too big and then they leave the preserve and start getting hit by cars. Or too many wolves and they start killing farmers' sheep. Culling helps avoid the wild swings in population.

I'd also point out that even with populations that are not in good shape (threatened or declining), selling super expensive hunting licenses can still make sense. If $50,000 = +3 rhinos, let the hunter kill the rhino. Of course there are perverse incentives and all sorts of practical difficulties, but I just want to make the point that it's possible for that sort of thing to be in the best interests of the species being conserved.

You didn't put much effort into your reply, but I'll respond:

"Minds differ greatly by culture" is something that is absolutely not obvious to most people (or most scientists). Cross-cultural psychological research is really only picking up steam recently.

Is the quote you gave politically correct? Depends on your culture :-D Is it true? Yes. But the thing you miss is that culture is like software...you can install new software, for the most part, in humans (teach them new ways of doing things). You're actually thinking about selection on the hardware, it seems.

I share your frustration with the fact that speaking the truth about the heritability of psychological traits is treated as heresy in lots of the developed world.

This is an ad hominem attack. The fact that you don't like his politics doesn't imply that he's wrong.

The Mosuo are _matrilineal_ (figure out which kinship group you're in based on Mom's line) but not _matriarchal_ (women hold most of the political power).

You'll occasionally see it argued that the Mosuo are matriarchal, but what's going on is many people are desperate to try to find an example of matriarchy, and the Mosuo are as close as it gets. So yeah, if you squint, stand on one leg, and crook your head a bit it sort of looks like they might be.

What's really the case is that every single human society we know of has been some flavor of patriarchal.

Minds can differ greatly by culture. For instance, some cultures only count one, two, many. Consider what a huge difference that would make to your cognition.

Lol. They weren't carving granite though, it was some very soft volcanic rock.

Again, I really appreciate you engaging and I'm happy for the opportunity to talk dane-pgp. I don't have much time atm so a little rushed.

>so you didn't have to trust the writer or editor of the article

They're linking to a lot of their other stuff, iirc, and selectively citing things that support the position they want to disseminate.

> It says that there are 3 types of laws which have the strongest level of scientific support for their success

We have to differentiate between effect size and statistical "significance" (poorly named). The effect sizes (how much of a difference you would expect to see) are small iirc. For instance, the "assault weapons" one, which gets the strongest level coloring, is I believe a 0.9% decrease in murders if it's the study I'm thinking of. The "statistical significance" is, roughly, how much you can believe the effect size you're measuring is real, NOT how important it is. So you can have a 0.9% reduction in murders that is very "significant". I don't know what their coloring is, but my guess is they're using statistical significance, maybe with blending in some effect size.

>I wonder, though, what it would take for me to convince you of anything, since you could just dismiss all scientific research as "coming from academia" and therefore tainted by political views that you don't like.

I am from academia! I actually don't have a guns axe to grind specifically, it's a "we're going to poison the social sciences via publication bias and p-hacking" axe. Basically, people produce evidence for what's politically palatable (not even what they actually want to believe), and it's a huge problem. Another one is sex and behavior. It's abundantly clear that there are huge differences in behavior between the sexes, and that this is driven by genetics, and we've known this on the basis of good scientific evidence for 100+ years. It's not ~~socially determined~~.

>"Importantly, however, where we conclude that evidence for a policy is weak, that does not mean that the policy is ineffective; the policy itself might well be quite effective."

> "Still, even relatively small effects of gun policies are important to the people and communities affected."

Let me translate: "Guys, we know this doesn't look good, but we promise...there's a way you can talk about it such that we can still keep claiming the things we want to believe are true in spite of the avalanche of evidence coming from researchers that also want to claim the things. Also, jk jk, don't punish us we want to keep our jobs :-) :-D "

One of the key problems here (and the reason for my axe) is that it's tough or impossible to do real research on important stuff like this right now.

> extending that disapproval to mature ad hominem arguments as well. ;-) I've enjoyed this :-)

Haha, you make a bunch of good points. Again, I really appreciate the good-natured engagement.

> The absolute size of the risk is what's relevant.

If you look at the whole sentence "...compared to things like...", I'm talking (or at least meant to be clear about talking) about the risk from shootings in general, not shootings of children in particular or school shootings:

> Shootings, especially mass shootings, are a very small problem compared to things like obesity, diabetes, heart disease, car crashes, and tiktok. If you're taking a vaguely utilitarian approach to public health policy it makes the most sense to do things like try to make Americans less fat, by a wide margin.

When I say "shootings", I mean the thing people intuitively understand it to be (Person A shoots Person B), and not suicides. If we include suicides, car crashes and "shootings" in the U.S. are ~ the same size. Suicides don't respond much to "no guns" as I recall...you get a couple percent reduction, but people mostly substitute (tall buildings, etc.). I happen to think it's cruel to prevent all suicides, and that some people who want to die should be allowed to do so with dignity, instead of alone and sad, but that's a digression.

Anyway: for adults, the relative risk is important since we're making decisions about allocating public health resources. There are on the order of a few hundred thousand "unscheduled" (preventable) deaths in the U.S. in a given year, and 12,000 gun murders. Get rid of guns, and charitably that number goes from 12,000 murders to 9,600 murders. But you could've saved a lot more lives working on the boring stuff, not to mention the astronomical amount of political capital you burned to get that done.

For kids, the the absolute risk is what's important since they basically don't die. If you're thinking about where to invest resources, invest not in making them die less but rather in making them into better, healthier people. Not to say investment in preventing childhood death should be exactly zero, I'm talking on the margin here (what do we do with the next dollar/minute/effort).

>even one school shooting is too many

...yeah, but it's not! Would you pay $100B to prevent the next one? Could've saved a lot of people from diabetes / heart disease with that money, instead of 8 kids. It's tragic, sure, and we're all tempted to say stuff like that for social reasons, but when you're really deciding how to allocate resources it's so, so rare that hedging doesn't make much sense. Remember, about double the risk of being struck and killed by lightning (a lot more people are struck and survive).

I don't have time right now to go in to the race thing, but I will point out that "Black" is ~14% of the U.S. population. Black people (90% men) commit more than 50% of homicides in the U.S., so they murder at ~7x the rate of non-Black Americans before correcting for things like which crimes get solved (correcting moves it to ~10x, they're better at getting away with murder since organized crime). If you're curious, this is where you should start (the UCR): https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-...

It's true, but just to clarify for people out there: "yard trees" bring less, and maybe nothing...if you can get someone out there. It's because there's usually only one or a few in that spot, they're difficult to get to (behind a house), and they're full of metal that will damage saw blades (like that clothes line pulley someone put up in 1930).

I don't know where you live, but a lot of municipalities make what your neighbors are doing illegal. Might want to check.

Young people basically just don't die much. Whether something is the leading cause or the 3rd cause is irrelevant. The absolute size of the risk is what's relevant.

Here's an example: Imagine you hear that the U.S. is 43rd in the world on math test scores among highschoolers, and China is 4th. ZOMG, A BIG DEAL!?!? Right? Well, that's what you'll hear in the media anyway.

Now, imagine if I told you that the mean U.S. score was 87, and the mean Chinese score was 89 out of 100. Does it still seem like a big deal?

Here's another point from another comment: This is a bit disingenuous I think. I worry people will infer that young children are at risk from random violence. That's not true.

That stat goes from 0-19 years of age, and the vast majority of deaths are in the older segment. Like everywhere in the United states, it's young Black men killing other young Black men as part of organized crime or over matters of honor.

If you're a parent and not participating in that world, you and your children have nothing to fear.

I appreciate you engaging! Really, I do.

I am sorry to say it, but Scientific American has become politicized trash under current leadership. It pains me because I remember it fondly from my childhood. The article you linked to doesn't even try to meet basic scientific standards.

Here is a good overview: https://www.rand.org/research/gun-policy/key-findings/what-s...

TL;DR: evidence coming from academia (which is extremely anti-gun) only weakly supports a very small subset of gun laws as working. Taken in the light of publication bias and the file drawer problem, this is a nothing.

You could be right. I hope not!

I think you're right about it being a bigger deal elsewhere, and you're right about legislative reactions. I have several thoughts:

1) The U.S. population is really big. 333M or so. New Zealand is 5M? So, inflate the base rate by 66x. When doing "intuitive statistics" about this sort of thing, people think at the country level and don't adjust for population. That will remove some of the effect, but even doing that there's going to be more school shootings in the U.S. per capita, of course.

2) Gun laws don't work. The best available work on this suggests a modest decrease in murder rates at best. People who want to murder will just use alternative means. There are plenty of ways to kill schoolchildren if someone decides to do it. Easy availability of guns probably does increase the rate and deadliness, but not by all that much. Something else is going on in the U.S. Media contagion is part of it, imho.

3) The extent to which people pay attention to astronomically unlikely things like school shootings really doesn't make any sense. As I estimate somewhere else in these comments, the likelihood of a child in the U.S. being shot and killed at school is about double that of being struck and killed by lightning. Though it's a tragic spectacle when it happens, it's not something worth worrying about. The irony is nobody gives a shit about stuff that might actually kill your kids, like drowning, car crashes, etc. People are so, so very nonchalant about things that are actually risky...

I mean, you might be right about Mexico! I don't know the details. The fact of the matter is that Mexico is the way it is, and that has effects on violence in the U.S.

You seem concerned with blame (normative statement), but I'm just making a descriptive statement (how it actually is right now).

As for foreign wars increasing violence domestically in the U.S., I don't buy that. Most of the murdering in the U.S. is some cultural thing to do with young black and brown men that we haven't figured out yet (because it's essentially impossible to talk about it, much less do research on it).

The rifles that Americans own themselves are just fine. Average quality is better than military, though the military does get features that are illegal for civilians (like full auto).

330,000,000 potentially armed people (or about 260,000,000 armed adults) is something to be reckoned with. A bunch of goat-herding peasants in Afghanistan fended off not one but both superpowers, and at that mostly with small arms. The fact that the U.S. chooses to arm its population is definitely relevant for the direction its politics goes. Whether that's good or bad, net...well, we never get to know because we never see the counterfactual situation.

Shootings, especially mass shootings, are a very small problem compared to things like obesity, diabetes, heart disease, car crashes, and tiktok. If you're taking a vaguely utilitarian approach to public health policy it makes the most sense to do things like try to make Americans less fat, by a wide margin.

> So the idea that a handful of untrained civilians with their Wal-Mart rifles could have prevented any of this is delusional fantasy.

The thing about the U.S. is it's not a handful. They have more guns in civilian hands than people. It would be very easy for U.S. civilians to arm literally everyone.

And, I should note, the quality of the rifles people own privately is as good as or better than what the military gets. The military does, however, get full-auto.

I take your point about the Nazis. I suppose if I had to point out a difference with respect to the U.S. today, it would be that the right to keep and bear arms is for everyone. Americans really don't discriminate (at least overtly), and that's a core tenet. The Jews in Nazi Germany obviously were or had to be disarmed before terrible things happened to them.

Heh, I think the level of dislike we're talking about here is different. You may not like them but it's nothing like being a Jew under the Third Reich or, perhaps, living in eastern Ukraine right now. The point at which "power comes from the barrel of a gun" becomes relevant someplace like the U.K. is a ways off for almost everyone living there, thankfully.

The worry is twofold: first, not having an armed populace now might allow things to get really bad in the future, whereas in the case where everyone's reasonably well-armed, or has the option to be, it wouldn't have gotten that bad in the first place. Second, if things are really bad at present, bad to the point that fighting seems reasonable, you want to be able to fight. Disarming now means you'll likely be disarmed in an uncertain future, too.

Our modern social democracies seem really stable, but from the perspective of a historian there is no reason to believe they are. They're infants. There's a war in Europe right now, and there are people alive who remember the Holocaust.