Also this week: The Dead of Winter, the economic inefficiency of holiday social obligations, comedy wildlife photos, the “piranha problem” in social psychology (and ecology?), uncited papers are rarer than you think, there’s no replication crisis in ethnography (and that’s a bad thing), and more. Lots of good stuff for your holiday reading pleasure.
A day late (because it’s about the solstice), but read it anyway: The Dead of Winter. Short, fascinating, and at the end deeply moving. The best thing you’ll read this month, I promise.
Ethnography has no replication crisis, because “the truth basis of ethnographies is basically ‘trust us’.” Very interesting read. (ht @noahpinion)
What fraction of scientific papers never get cited, at least after some reasonable period of time? The journal Nature investigates and finds that the true fraction is much less than the 50% number you’ve probably heard.
Deborah Mayo critiques the argument for redefining statistical significance.
This economics prof gave her final exam via Twitter. Yes, really. You could totally do the same thing in other classes, though whether it was a good way to test mastery of the material would of course depend on the class. (ht @noahpinion)
A little while back, prominent social psychologist Dana Carney publicly declared that she had lost confidence in her high-profile work with Amy Cuddy on power posing. Now the Loss of Confidence project is encouraging other psychologists to come forward with their own stories of loss of confidence in their own results, with the idea of writing a paper on which everyone will be a co-author. The goal is to destigmatize the public admission of error. Related posts from Brian and Meghan. (ht Retraction Watch)
Roger Peng on the challenges of teaching data science, as opposed to teaching traditional statistical topics that are (i) broadly applicable, and (ii) have formal theory.
The co-author of a now-retracted Science paper claiming that microplastics negatively affect larval fish ecology, who was found by Uppsala University to have intentionally falsified the results, is now under investigation regarding doctoral work. (Note that I’m just passing on this link; I haven’t read it because I don’t subscribe to the linked newspaper.) (ht Retraction Watch)
Andrew Gelman on the “piranha problem” in social psychology: the many studies claiming huge, repeatable effects of seemingly-minor factors on human behavior can’t all be true, because if they were they’d interfere with one another and so wouldn’t show up as huge and repeatable in the data. Gelman’s talked about this before, but I thought this was a particularly clear statement (though I’d also like to see some simulated data to make sure I’ve fully understood the argument). Am thinking about if/how this applies to ecology.
The ABC conjecture has still not been proved. Interesting insights into the anthropology and sociology of pure mathematics. (ht @matt_levine)
A profile of mathematician Jim Simons, hedge fund billionaire, founder of a foundation that supports basic research in various fields (including ecology), and now founder of an institute devoted to computational biology and physics. He either is or is close to becoming the biggest private funder of basic research in the US. (ht @matt_levine)
On the marginal position of research on X within discipline A. Or, when is it valid to criticize an individual researcher for not giving sufficient attention to marginalized topic X, out of all the many marginalized topics to which more attention could and should be given by the discipline as a whole?
And finally, Caroline Tucker’s annual holiday caRd. 🙂