Also this week: low-risk, low-reward publication strategies among women social scientists, whatever happened to 90s environmentalism, Andrew Gelman vs. Judea Pearl, Twitter vs. harassment (but not in the way you think), scholarly disciplines as conversations, and more. Lots of good stuff this week! (for some value of “good” and the usual values of “lots”, “stuff”, and “this week”)
The world debates Plan S, the European-backed effort to mandate that all scientific journals immediately switch to author-pays open access.
Ductor et al. (unreviewed preprint) find that women economists and women sociologists consistently have lower research output than men on average by various measures, even after controlling for author experience and subfield. This is associated with gender differences in co-authorship networks: women have fewer but more experienced collaborators than men on average. Ductor et al. interpret their findings as indicating low-risk, low-reward choices by women authors: women are more likely than men to stick with a few experienced collaborators they know well, and to target lower-tier journals. Ductor et al. speculate on various possible explanations for these findings. They focus on the possibility that women are trying to mitigate/avoid the risk (or perceived risk) of receiving gender-biased peer reviews and editorial decisions from leading journals. I’d be curious to see a similar analysis for ecology. (ht @noahpinion)
Science reports data on gender balance in its publication process. The statistical analysis kind of falls in the uncanny valley between data exploration and rigor for me, but the main result is clear enough: women are a minority among first authors of Science papers, but acceptance rates for Science papers first-authored by women and men are pretty similar. Over the last decade, there’s a trend in which acceptance rates for women first authors have gone from slightly lower to slightly higher than those for men. The proportions of women among authors of Perspective pieces, and quote sources in news stories, are growing steadily but still low. All this broadly aligns with similar analyses I’ve seen from other journals, such as Functional Ecology.
A couple of nuggets on determinants of undergraduate academic success, from a session at the recent ASSA meeting (ht @noahpinion). Both results are from unreviewed preprints which I haven’t read; just passing them along in case you want to dig in yourself. First, in a big randomized experiment, an easy-to-implement, low-cost, scalable way to encourage students to plan ahead and study more got high engagement. Too bad it utterly failed to have any effect on academic performance or retention. Second, there is an easy-to-implement, scalable way to reduce academic performance: let students join fraternities and sororities. (From Meghan: Paying for the Party, is a book by social scientists who study college culture, finding that the party culture of fraternities and sororities helps maintain inequality. It’s a very depressing, but also important, read.)
UC Irvine professor Kathleen Treseder, who chairs the biology department there, took to Twitter to get UC Irvine to protect her from harassment by Richard Symanski, an instructor in the department. Be warned that the linked article includes disturbing, misogynist passages from Symanski’s self-published autobiography. (ht Small Pond Science)
Andrew Gelman reviews Judea Pearl and Dana Mackenzie’s new book on causal inference. The review is fairly critical. Hoping Jim Grace will chime in with his thoughts, as he’s one ecologist more familiar with Pearl’s ideas than I am.
Kieran Healy on “conversational disciplines“. Here’s the conclusion:
Disciplines fail or break down in the same way that conversations fail or break down: for lack of conversation partners, for lack of interest in a topic, or for chronic lack of agreement on what counts as a useful contribution. While some kinds of failure just lead to embarrassing lulls, the latter sort of failure is potentially destructive of the conversation itself.
Includes the interesting suggestion that a discipline can remain coherent even in the face of serious disagreement about what questions are most worth asking and what the answers are, so long as there is widespread agreement on methodology. Go read it, then think about whether/how it applies to our many discussions of whether ecology is making progress or is even a single coherent discipline.
What ever happened to 90s environmentalism? I don’t know enough to fully evaluate the suggested answers, but the question is very interesting. (ht Marginal Revolution)
I’m slightly late to this, but here’s Timothy Taylor looking back on 50 years since Garrett Hardin’s “The Tragedy of the Commons”.
Teaching an intro stats course? Here’s your new go-to illustration for why “haphazard” is not “random”. (ht @noahpinion)
All of William Wimsatt’s work in philosophy of science condensed into one sentence (the second image in the linked tweet). 🙂 I say this fondly, by the way; I’m a huge fan of Wimsatt’s work.
Biostatistical baking (ht @kjhealy). 🙂
I assume this was produced in knitr. #sorrynotsorry (ht @kjhealy) 🙂
Too many tabs open, 18th century edition. (ht @dandrezner) 🙂