Also this week: igniting the ESA meeting, improving retention of female students in male-dominated fields, scientific aliases, 3D printed mouse penises, stupid things we refuse to link to, and more.
In an old comment thread we discussed the interesting idea of an arXiv overlay journal. The way it works is that anyone who publishes a unreviewed preprint in discrete analysis on arXiv can ask the “overlay journal” Discrete Analysis to review it. They have a board of editors who arrange the usual sort of peer review, in their case evaluating preprints on the basis of interest/importance/etc., not just technical soundness. If the preprint is accepted, the journal “publishes” it by linking to the preprint DOI. The idea is to decouple two functions of traditional journals that no longer need to be coupled: evaluation of papers, and publication/distribution of papers. Thereby dramatically lowering the costs of open access publishing, since posting to arXiv is free to authors. Now somebody’s doing something similar for evolutionary biology,
though with significant wrinkles that to my mind subtract rather than add value. In particular, the editors (called “recommenders”) find their own papers to recommend rather than relying on author submissions, and they can recommend published papers as well as preprints. Which means there’s a fair bit of overlap here with Faculty of 1000, not to mention “people using social media to recommend stuff they liked”. I think the recommenders are likely to end up mostly recommending papers that would’ve come to readers’ attention anyway, thereby making the recommendations largely redundant with existing ways of filtering the literature. The initiative advises recommenders to prioritize unreviewed preprints, but so far that’s not what they’re doing. The early recommendations are dominated by papers from high profile journals like PNAS, Current Biology, and Plos Biology. The one preprint I saw was from the lab of a well-known PI. This highlights a major, often-overlooked function of conventional peer review of mss submitted by authors: it’s an attention redistribution mechanism, not an attention concentration mechanism. Arguably, the world has plenty of methods of attention concentration and needs more methods of attention redistribution. But it’s early days and time will tell, I think it’s great that people are experimenting with new ways of doing things. CORRECTION: I got much of this wrong, my apologies, and thank you to Ruth Hufbauer for correcting me in the comments. You should click through and read Ruth’s comments, but the short version is that this new initiative is much closer to that arXiv overlay journal than I had initially realized. In particular, the recommenders are doing full peer reviews of preprints and papers submitted for their consideration by the authors. And reviewing already-published papers is just a way for them to help get the word out about this new initiative. I think it’s a very interesting new initiative and look forward to seeing how it develops. (ht Andrew Gelman, who see this as an improvement on related initiatives: “ArXiv is too unstructured, NBER is a restrictive club, traditional journals are too slow and waste most of their effort on the worst papers, twitter’s a joke, and blogs don’t scale.”)
Terry McGlynn wants many more Ignite sessions at the ESA meeting to cut down on the number of parallel sessions and boring talks, and foster more discussion. I think the current mix of talks and session types is fairly optimal, though I’d like to see the ESA go back to the 15 minute time slots they used to have rather than the current 20 minute slots. Thoughts?
Do university administrators have the wrong incentives? I agree that central administrators have overly-strong incentives to start new initiatives that create work for faculty and frontline staff, and that require hiring of additional administrators. But in my experience these initiatives don’t generally take the form of real estate projects.
I’m a bit late to this: Rich Lenski invites you to vote on whether he should modify his famous Long-Term Evolution Experiment to include a second treatment (transferring a smaller fraction of the population to fresh culture medium each day, so as to get more generations per day). Includes a fun quiz as what effect you’d expect the second treatment to have. Follow-up post here, in which Rich discusses his own seemingly-frivolous-but-actually-profound way of answering his own quiz question. Related: my interview with Rich from 2015.
This week in Things I Refuse To Link To Because They’re Stupid: As you’ve probably heard if you’re on Twitter, a couple of philosophers got a fake paper consisting of satirical nonsense and penis jokes published in a social science journal. Another Sokal Hoax, right? Except the journal in question is a predatory author-pays open access journal that will publish just about anything. And except that the paper was first rejected by another extremely obscure journal, which then cascaded it to the predatory journal. It’s like the authors were trying to shoot fish in a barrel in order to prove that fishing is bad, except that they shot at the fish and missed, so instead they bought some past-date frozen fish sticks from a dodgy supermarket, ate the fish sticks, and then claimed that was somehow the same as shooting fish in a barrel, thereby proving that fishing is bad. Well done to all concerned, including anybody who retweeted or linked to this stupidity approvingly, for making the world a little worse. At least Steven Pinker reversed his initial stance on this.
This has nothing to do with any of our usual concerns around these parts, but it will make you laugh and warm your heart: a small illustration of what a great guy Roger Moore was.
Hotel room aliases of statisticians. So, if you were going to register at your ESA meeting hotel under an alias, what would it be? I’m vacillating between “Price, G.”, “Gause, G.”, and “Moyer, J.” Minus 10 Internet Points for lack of imagination if you picked “Darwin, C.” I predict Meghan will choose “Magna, D.” 😉
And finally, Stephen Heard named a newly discovered moth after Dynamic Ecology! Sort of. Suggest Latin names for the Dynamic Ecology moth in the comments. 😉
Rich Lenski gave the PhD commencement address at UNC-Chapel Hill. It’s full of lots of great thoughts for new PhD grads (and lots for the rest of us, too). It talks about the role of chance in life and science, about his start as an ecologist, and about the lucky breaks that got him started on his now-classic long term evolution experiment (and some good advice that kept it going).
New data from Britain’s Higher Education Statistics Agency (HESA) shows that the number of students who have withdrawn from college due to mental health issues is increasing rapidly, up 210% in 2014-2015 compared to 2009-2010 (the first year for which they have data). That article (by The Guardian) showed increases in requests for counseling, too, which, in at least some places, has led to increased wait times.
Ed Yong had a piece on the problems scientists sometimes have going through airport security, due to carrying things like 3D printed mouse penises in their carry-on. (I also love that he notes that saying “you don’t really understand something if you can’t explain it to your grandmother, a barmaid, a six-year-old” is sexist and ageist. Yes!)
Ed Yong also had a piece on the influence of female mentors on engineering undergrads. He writes, “female engineering undergraduates who are paired with a female mentor felt more motivated, more self-assured, and less anxious than those who had either no mentor or a male one. They were less likely to drop out of their courses, and keener to look for engineering jobs after they graduated.” The article notes that the effect is due to female engineering students with female mentors retaining their confidence, whereas those with male mentors lose it. As the article says, “You can’t be what you can’t see.” (Jeremy adds: Hey, I was thinking of linking to this! But since Meghan beat me to it I’ll just add my comments here. I had a look at the underlying PNAS paper. It seems well-designed and the main results are all as Meghan describes (though I don’t understand why the figures have no error bars…) But I wanted to touch on the one result that runs counter to the general thrust of the paper, not because I think it’s the most important result but because it puzzles me a little and I want to hear what others think of it. Women assigned female mentors or no mentors saw their GPAs decline over time; women assigned male mentors experienced no GPA decline. The GPA effects weren’t massive, but they weren’t trivially small either, at least to my eyes. And they occurred even though mentees assigned to male vs. female mentors reported no differences in how often they met with their mentors or how often they discussed course material and study skills. The authors downplay this result because GPA doesn’t translate into satisfaction with or retention in engineering, which is fair enough. But presumably, there are other reasons to care about GPA. I found myself wondering if all mentors could be given different training so that their mentees could get the best of both worlds–mentoring that improves their satisfaction with engineering, their retention in engineering, and their performance in their coursework.)
Here’s a rubric that you can use to assess your syllabus to see how student- and learning-centered it is. I found it useful, especially the parts on pages 3-6.