I've had some inspiring discussions with very bright people and given and received collegial support over the last couple of days, and that makes me happy. But then I've also heard of people acting like jerks. Why do people have to act like jerks?

I fully agree that a fully standardized effect size can often be a bad criterion for dismissing an effect. But somehow, instead of choosing more sensible metrics for evaluation, psychologists tend to conclude that hence any “significant” effect is potentially very important.

I've sometimes received requests that I shouldn't put confidence intervals on estimates or even report response rates for surveys. I've declined politely.

I feel this is a common sentiment among people who are not trained in statistics. Uncertainty is weird and hard to think about, ignoring it makes it easier to digest information (even if it causes a false sense of certainty).

Not exactly that, but check out priorsense: github.com/n-kall/prior...

Except maybe for the better.

I don't know why I find this so funny, but I nevertheless do.

chains = 0 :D That's clever!

With brms models? I'd be interested in hearing how you did that.

My models run so quickly and I've had some problems with {future} in the past, so I usually just run each model with 4 parallel chains instead of trying to fit all the models in parallel.

Here's a piece of code I've used to fit multiple bernoulli models to nested data frames (can't remember where I picked the techniques up). You could probably replace map with future_map like @ajordannafa.com does and maybe also skip running the first model somehow. Code in alt-text.

Only randomized controlled trials are gold standard science, so epidemiologists should preferably start multiple epidemics in random cities.

Exactly! I have a friend who lived through a similar transformation in Finland, and they said the final coup de grâce to the polite second person plural was the proliferation of email, because it meant you were more and more in contact with people whose social status you didn't have any idea about.

Maybe this is a good time to mention what my mind plays everytime I see @konsta.happonen.eu

Ha! I am inevitable.

In Sweden, abandoning titles was actually a conscious effort in the sixties, spearheaded by people working in Stockholm.

en.wikipedia.org/wiki/Du-refo...

Wait, does this mean that non-Euclidian cities, the scariest things according to Lovecraft, are real?

It's so weird, and especially so with flat "uninformative" priors on the logit scale, which assign almost all of the probability mass to zero and one.