Bears, guns, and Battlestar Galactica

José Duarte, PhD

UPDATE: I heard back from Politifact the next day and they partly corrected their article. See my addendum at the bottom of this essay...

Politics is the worst. It breeds malice and it makes people stupid. I mean incredibly, implausibly stupid.

Betsy DeVos is President Trump’s nominee for Secretary of Education. Last week, she testified before the Senate as part of the nomination process. Someone asked her about guns and… boom! Within 24 hours America’s average IQ tanked as ThinkProgress and Jezebel reported DeVos’ response while stripping the very specific context to which she was responding. Politifact even doctored DeVos’ quote, which I’ve never caught them doing before, and reported a wildly false statistic, which is more typical of them (let’s clock how long it takes them to run a correction).

This whole thing was a ridiculous waste of everyone’s time and it triggered a disturbing parade of malice and vice that people really should want to avoid. We only get one life, people — do you really want to spend it working up hatred and mockery for half the country every few days because of a bogus story you read on Jezebel, ThinkProgress or Politifact? Is this the life you want to live? Do you want to be so gullible and pushed around by bad statistics? America needs to be a better place than this.

Here’s what happened: Before the hearing, DeVos met with Senator Mike Enzi of Wyoming. He mentioned that Wapiti Elementary School, a tiny rural school near Yellowstone National Park, had put up a fence to keep out grizzly bears, which had been a problem for the school. In fact, one former problem bear is now a pelt hanging on the wall outside the principal’s office (as of 2005):

“That one,” says Wapiti teacher Angie Terri, “was a problem bear. It was hanging around the school, and then it started hanging around the post office just down the street.”

During her hearing, Senator Chris Murphy wanted DeVos to say “definitively, that guns shouldn’t be in schools”. DeVos declined to commit to such a simplistic position, pointing to the context of some rural schools like Wapiti’s:

“I will refer back to Senator Enzi and the school he was talking about in Wyoming … I would imagine that there is probably a gun in the schools to protect from potential grizzlies.”

This is a pretty mundane topic for rural communities. But undereducated urban liberals went wild, spun the story, deleted the context, screwed up the stats, and laughed at stupid Republicans in the usual venues. Politifact swung into action to “fact-check” an explicitly speculative claim that started with “I would imagine…” and was further qualified by probably — and they deleted the word probably from her quote. They deleted it. No ellipsis in its place— it’s just gone:

ThinkProgress provides the full quote, but they never tell their readers what she’s talking about. They don’t mention the school in Wapiti that had a bear problem and put up a fence, or that DeVos was specifically referring to that school and its past bear issues. What kind of scam is this? Does ThinkProgress have a code of ethics up somewhere? They should.

Jezebel is in Jezebel form and of course doesn’t tell its readers what DeVos was referring to. I mean, why would they? They’re Jezebel. They just make their readers think this crazy lady is worried about grizzlies. The commenters talk about how evil she is, about how this story makes them cry, how she’ll destroy public education, etc.

I’m worried about how many Americans are living in a comic book at this point, with a wildly distorted perception of reality, particularly the reality of how other humans beings’ minds and motives work — Jezebel is Exhibit A regarding this concern.

I see a problem here with people having serious deficits in their social cognition. They’re not accurately modeling the perspectives and motives of a large number of other human beings. In particular, urban leftists seem to not understand people who aren’t urban leftists. Since most Americans aren’t urban leftists, this is a recurrent and somewhat frustrating deficit to bump into. We see it a lot in academic social psychology, which consists almost entirely of white urban leftists who embed their cultural assumptions and ideology into their research in ways that often destroy its validity. It looks like we have a similar problem in the media, where this urban leftist culture is overrepresented.

Now, I know some of you will argue that it’s okay for the media to have this bias, because you think leftism is a true and correct ideology, or that smart people will lean left on the merits. That’s certainly possible, but you should be smart enough to automatically asterisk such an explanation given how convenient it is, and how you might be biased and motivated to want it to be true. At the very least, you’ll want to rigorously examine this hypothesis, and always be on the lookout for contrary evidence — you should cultivate this habit in yourself if you want to live with integrity and be as reality-aligned as possible.

In any case, stories like this one should curb your enthusiasm for the notion that reality has a liberal bias. Let’s note the following:

  1. Leftist media actually called the Wapiti school district to ask if the school had a gun, apparently on the assumption that the locals would tell them if they did. Fascinating.
  2. A Wapiti school official reports that some reporters calling about this story had trouble believing that there could be grizzlies near a school. This is also a fascinating belief for people to have about the world. Here’s what a rural school looks like from above (the red marker is on the school, which is tiny):

3. The media thinks that it matters whether a small school in rural America “has” a gun, as in a firearm stowed in an office somewhere. This suggests a profound lack of insight into rural life. It doesn’t matter if the school has its own gun — the adults can just get one from their cars. Here’s what a rural school looks like up close:

(As it happens, both photos are of the Wapiti Elementary School…)

Bears, bullets, and BS

Various media seem eager to take this opportunity to report that guns don’t stop bears. I’m going to fast-forward a bit here — these people are very stupid.

These same media also argue that a school would be better able to fend off grizzlies with bear spray than with a firearm. This is even more stupid, and the fact that I’m willing to continue this essay at all, is, in my mind, an act of remarkable charity.


Politifact screwed up (or lied, which is how they always characterize falsity — they don’t have an error category in their rubric). They said:

A study of bear attacks in Alaska found that in 27 percent of cases involving guns, the person couldn’t get their gun ready fast enough to shoot the bear…

False. It was 27 percent of the subset of cases where a firearm was ineffective, which were a minority of cases, not 27 percent of cases involving firearms. See paragraph 3 of the Results section, page 3.

Note that since firearms were extremely effective in defending against aggressive bears — 84 or 76 percent of the cases in which they were used, depending on type of firearm — the correct figure for Politifact’s claim is somewhere around 5 percent. (Multiply 27 percent by 16–24 percent)

Joe pause: Politifact is unbelievably incompetent. It stuns me how incompetent and partisan they are. Can anyone confirm that they only hire leftists? From what I’ve read, they don’t employ non-leftists, which would laughably undermine any claims of seriousness, and would explain their low standards — people get lazy when everyone agrees with them. I’ll report more on Politifact soon. There seem to be a few fact-checking scams out there that only hire leftists — that’s nuts, and it reminds me of the Facebook trending news team. (If you know of any conservative or libertarian fact-checkers, please send me links.)

Continuing…Something happened to the Washington Post when I wasn’t looking — they don’t seem to do journalism anymore. Their headline: Actually, Betsy DeVos, guns are a pretty lousy way to ward off grizzlies

Actually, WaPo, guns are extremely effective at warding off grizzlies. As noted above, the source you cite reports that they stopped 84 and 76 percent of attacks in which they were used, depending on the type of firearm. The Post never discloses this simple reality. Instead, they report peripheral statistics:

“It examined 269 bear-human encounters in Alaska between 1883 and 2009 and found that people carrying guns suffered the same injury rates whether or not they used their weapons.”

What do you think that means? I have no idea. Now we’re only talking about relative injury rates, which is one variable out of many. You would need to dig into the data and unpack this. It could mean that injury rates were very low in both cases — a floor effect. It doesn’t tell you how effective guns were, especially in this century. It doesn’t tell you what the people who had guns but didn’t use them did instead — did they run? Is running the best strategy? (My understanding is that it isn’t.) In any case, we’d have to dig in to find out. Because the dataset spans more than 100 years, the main effect could be distorted by cases from a specific period, and that period might not be the present. We don’t even know what counted as an “injury” for this study. These are all things you’d have to know, but the Post is too lazy and partisan to do the work needed here.


  • On a school depending on bear spray over a firearm… Bear spray is a personal defense weapon. It’s pretty good at that — evidently it’s even more effective than firearms, though once you’re up in 80+ percent effectiveness territory, I wouldn’t worry about purported differences. A school is not a personal defense scenario — it’s not a hiker in the woods. If you’ve got a bear on school grounds, you’d have to approach it to use bear spray, which is an incredibly bad idea. For a facility that is worried about bears, you need an area defense weapon, like a rifle, not a personal defense weapon. Well, “need” is arguable. A fence might do the job. And if it fails you can probably just get the kids inside and call the bear authorities or whatever. But if you were going to have an active defense — in the office, or in your car— a rifle or shotgun is much, much smarter than a spray. The spray-is-better-for-a-school idea was too stupid to be on major news sites in an advanced civilization — this is the kind of stupid that gets people killed.
  • US News is the only media I’ve seen that reported that the school district would like to have a gun in the school:

“I would support the school board’s identifying employees and training [them] and providing weapons, just for general protection in the school system,” he says. “These are isolated schools. If something happens in those areas, it’s not like the response time is very quick.”

This is quite reasonable. See the satellite photo up above. Rural Americans live in a very different, and less neurotic, world than the urban left. Firearms are a good example — many of the incredibly bad statistics that leftists rely on when arguing for gun control are driven by inner city violence, or by a cognitive bias known as the accessibility heuristic. These statistics have nothing to do with rural America, the NRA crowd, etc., so it makes sense that they don’t fret over firearms. The reporters on this ridiculous story don’t seem to realize that almost everyone has firearms in rural America — the school doesn’t need to have a firearm in order to have one.

José L. Duarte recently earned a PhD in Social Psychology at Arizona State University, and is about to launch the Valid Science Center. You can email him at

Joe is broadly pro-bear, and would never want to see one killed.

Addendum: Politifact correction

Politifact’s original reads:

A study of bear attacks in Alaska found that in 27 percent of cases involving guns, the person couldn’t get their gun ready fast enough to shoot the bear…

The correct number is actually 4 to 6 percent. The 27 percent figure was actually the percentage of cases where guns had been unsuccessful in defending against an aggressive bear, which was a small minority of cases, not the percentage of all cases involving guns.

Politifact “corrected” the above to:

A study of bear attacks in Alaska found that in 27 percent of cases involving guns that failed, the person couldn’t get their gun ready fast enough to shoot the bear…

So “27 percent of cases involving guns” became “27 percent of cases involving guns that failed”, which is not just clumsy writing, but easy to misinterpret as being about guns that failed — like they physically failed, which is not what this was about. It would be clearer to leave the text alone and just correct the number to 4–6 percent.

José Duarte, PhD

Written by

Social Psychologist and advocate for scientific validity. I research the psychology of envy. I also develop new theory and tools for methodological validity.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade