Also this week: old school blogging, fake anteater news, an easy way to make multipanel figures in ggplot2, Canadian higher education is weird, tweetstorms vs blog posts, the last man who knew everything (?), the biology of superheroes, Wittgenstein vs. emoji, and more.
This week in old school blogging: Florian Hartig wrote a post pushing back against my post arguing that technical statistical errors don’t much slow the progress of ecology. I think Florian’s is the strongest counter-argument against my post.
This week in new school blogging (and writing): a good tweet thread can work well as an essay outline. Inspired by that tweet, the example of economist Noah Smith’s many excellent tweet threads on political and economic topics, and the example of Meghan’s recent to-the-point blog post, I may experiment with writing some of my blog posts more like tweet threads. Note that I’m not planning to do actual tweet threads, at least not with any regularity, because I still find Twitter to be a poor medium for discussion of tweet threads (or of anything else).
“If climate change is real, how come coastal real estate prices aren’t crashing?” Here’s the very good answer. I learned something about how to think about the economic effects of climate change.
Political scientist Samara Klar on why authors from historically-underrepresented groups might be underrepresented in top polisci journals without editors or reviewers necessarily being biased against them. I wonder if the broader point generalizes to STEM fields, even though not all the details do (e.g., in STEM fields it’s rare for someone to seek out feedback on a single draft ms from many conference and seminar audiences over a period of years). (ht @dandrezner)
A possible example of the Matthew Effect in science: early career Dutch researchers whose grant applications to two Dutch research councils scored barely high enough to be funded went on to obtain much more future funding, and were much more likely to become professors, than early career researchers whose applications fell just short of the funding cut-off. That’s in part because applicants who didn’t get funded were less likely to apply for funding from those two councils in future. And it’s even though the two groups of applicants went on to publish at similar rates and with similar impact as measured by crude indices like the h-index. I wonder if part of what’s going on is that early career researchers who are unsuccessful at getting money from those two Dutch research councils subsequently turn to other funding sources. In which case this wouldn’t actually be a clear-cut example of the Matthew Effect.
An article on that old chestnut, who was the last person to know everything that was known at the time? I’d say Leibniz, but I really have no idea. The linked article makes an unconvincing-but-still-interesting case for a much later and more obscure figure. The case is unconvincing because he seems not to have written on much of the leading-edge science, mathematics, and engineering of his day. (ht Marginal Revolution)
Via Andrew Gelman, an interesting-looking article (which I haven’t yet read; it’s on my list) on the neglected roles of “diagnosticity” and a priori theorizing in producing strong science. From psychology, but sounds like it might apply to ecology as well.
Over at Small Pond Science, Terry McGlynn and I agreed that “I and/or my students will start a blog about our research” is almost certainly a bad “broader impact” to propose in your NSF grant. A blog about your research is highly unlikely to reach an audience, and just having your students write a blog post or two is not quality professional development for them.
This is useless to me since I don’t use ggplot2, but it may be of interest to some of you. The ‘patchwork‘ package makes it easy to arrange your ggplot2 plots into a multipanel grid. Related: Meghan’s old post on how people make multipanel figures. (ht @kjhealy)
The Wildlife Photographer of the Year competition has once again had to disqualify a prizewinning entrant because it was faked. The photographer took a stuffed anteater from the entrance to the nature reserve in which he was shooting, placed it in his shot, and then claimed it was a live anteater. (ht my dad)
Unusual features of Canadian higher education in a global context. Makes the very interesting point that, because big research universities are the norm in Canada, many students can attend what are widely regarded as “top” institutions. So status anxiety is much lower than in, say, the US or UK. Note that I don’t think this has any policy implications for other countries. I certainly don’t think the US should, say, get rid of its top liberal arts colleges to try to be more like Canada! I just found it interesting.
Evolutionary biologist Shane Campbell-Staton has a podcast called The Biology of Superheroes, co-hosted by Arien Darby, a marketing manager at Warner Brothers. Link goes to an interview with Shane Campbell-Staton. I see this as part of a long tradition of talking about academic topics by linking them to stuff people care about in their everyday lives. I’m thinking of everything from books on “the chemistry of cooking” to the class on “philosophy of sports” that my undergraduate college used to offer.
How many studies would you typically need to run to get a significant effect at the 0.05 level if you were studying a population in which the true effect is zero, and you ran studies sequentially and stopped as soon as one came out significant? The answer might surprise you.
Overly honest faculty job ad. (ht Marginal Revolution)
Andrea Tang speaks for every professor who saw Avengers: Infinity War. 🙂
This week in “clickbait headlines I fell for”: the claim that Wittgenstein invented emoji. (ht @noahpinion)
And finally, click here if you want to know what nickname my department’s graduate students have had for me for many years. Without me realizing it. 🙂