Also this week: Meghan quoted in Nature, the role of universities in politically tumultuous times, “gotcha bias”, and more.
I’m a bit late to this one, but better late than never because I suspect it will be of great interest to many of you. Writing in Ecosphere (open access), Hampton & Labou use the raw data from the 2013 NSF Survey of Doctoral Recipients to quantify the career paths of recent US ecology PhDs. The headline results (but do click through, the whole paper is well worth your time):
- Involuntary unemployment of ecology PhD holders is low (3.3%) and job satisfaction is high. Those results don’t change if you restrict attention to post-2000 PhD recipients. My comment: this should come as no surprise. It’s roughly in line with the involuntary unemployment rate for all US PhD holders.
- Only 5.6% of those employed report holding a job unrelated to ecology. My comment: interesting, I’d have guessed it would be higher.
- About 40% of post-2000 ecology PhDs are employed as tenure-track faculty (I got that number from eyeballing Fig. 2). My comments: I’m surprised it’s as high as 40%, I’d have guessed lower. Hampton & Labou focus on faculty hires at PhD-granting unis rather than at all colleges and universities, a focus with which I would quibble. But that is just quibbling.
- They find that gender balance of ecology faculty is improving, though they focus on more time-aggregated data than I’ve done in my blogging on this. So if you didn’t know, recently-hired N. American tenure-track ecology faculty are 57% women, slightly higher than their proportion among post-2000 ecology PhD recipients (52.3% in Hampton & Labou’s data) and their proportion among ecology postdocs (46.2% in Hampton & Labou’s data).
- About 92% of post-2000 ecology PhD recipients identify as white. My comment: One could ask why the gender balance of US academic ecology has shifted steadily towards equitability for many years while racial diversity hasn’t moved nearly as much.
A review of The Wizard and the Prophet, a history of mid-20th century environmentalism. Traces the roots of the ongoing conflict between “prophets” who think we need to cut back to avoid global catastrophe, and “wizards” who think that technological innovation can save us. This was already on my reading list for this year, now I’m moving it up the list. Sounds like a good companion piece to The Bet, which I reviewed here. Here’s a quote from The Wizard and the Prophet, from which I learned something I didn’t know:
So ineradicable was the elitist mark on conservation that for decades afterward many on the left scoffed at ecological issues as right-wing distractions. As late as 1970, the radical Students for a Democratic Society protested the first Earth Day as Wall Street flimflam meant to divert public attention from class warfare and the Vietnam War; left-wing journalist I.F. Stone called the nationwide marches a “snow job.”
Edward Levi, then-President of the University of Chicago, on the role of the university in politically tumultuous times. Eloquent, and very timely–even though it’s from 1967.
A bit late to this, but Meghan was quoted in Nature last month with her top academic career tips: learn the power of yes, and take care of yourself.
An interview with Shahid Naeem about Naeem et al. 1994, the paper that can take much of the credit for kicking off interest in biodiversity-ecosystem function relationships. Lots of interesting and entertaining backstory here. Mostly familiar to me, but probably not to many of you. (Also, is my memory going? Have I linked to this before?)
Data on the gender balance of economics students from high school through graduate school. The male skew of economics PhD recipients has its roots in or before high school, but doesn’t get any worse (or better) from the undergraduate degree recipient stage to the PhD recipient stage. (ht Economist’s View)
An unreviewed preprint reporting survey data from political science, suggesting that preregistered replications are likely to be subject to publication biases favoring statistically-significant results, and favoring results that overturn previously-published results (“gotcha bias”). Some of the reported effect sizes are quite small, and survey data about hypothetical papers may not predict evaluations of actual papers very well. But the results support my priors so of course I’m going to share them. 🙂 (ht @kjhealy)
Thanks, Jeremy – I’m so glad others are finding the paper interesting, and I hope it will be useful. A quick response to your comments… the paper addressed “the apprenticeship model” – faculty training grad students to take their jobs (tenure-track faculty in research universities). So that is why we split out the research universities at various times, and report the ~20% in tenure track at research universities. Many are at colleges and universities with a much higher teaching expectation, and many are in non-tenure-track positions that are research, teaching, and administration. While specifics vary among all these positions, in general it points to the need for diverse training that prepares new PhDs to be competitive for diverse positions and then hit the ground running when they get there.
All very fair points Stephanie. My quibbling mostly just reflects my own approach and experience. I don’t think of myself as training my graduate students to take my job, and AFAIK I don’t think my colleagues here at Calgary think of themselves as following that model either. Anecdotally, I do think it’d be fair to say that here at Calgary, and elsewhere, we do a better job of preparing students for *academic* careers than non-academic ones (though I think and hope that here at Calgary we’ve improved on that front in the past couple of years). I don’t think the gap between how well we prepare students for research-focused vs. teaching-focused academic careers is nearly as big. But that’s just a subjective impression and I’m sure others’ mileage may vary on that. And again, this is just quibbling over emphasis, I completely agree with your general point about the need for diverse training for diverse positions, and that’s the important point here.
In case any readers are interested, I’ll use this comment as a shameless excuse to re-up some of our old posts on non-academic (and academic non-faculty) careers for ecologists 🙂 :
https://dynamicecology.wordpress.com/2013/05/22/guest-post-on-having-the-courage-to-build-your-own-non-academic-career-path/ (a contender for our best post ever)
https://dynamicecology.wordpress.com/2016/12/05/helping-grad-students-pursue-non-academic-careers-advice-from-anne-krook/ (lots of great points and suggestions in this one)
And this post has links to all our guest posts from ecologists in various non-faculty careers: https://dynamicecology.wordpress.com/2017/09/16/were-looking-for-guest-posts-on-non-academic-careers-for-ecologists/
And for those faculty and prospective faculty who want to improve their teaching, if you search our archives you’ll find lots of great advice from Meghan, and some from Brian and I as well.
The “Reflections on Papers Past” site is so interesting! So much read between the lines in the interview, and so cool to see the context and the people (i.e. humans) behind the papers. I bet it is an especially good read for young scientists.
Thanks, Jeremy, for calling my attention to that preprint from political science. Though imperfect as you noted, I liked the authors’ attempt to assess researchers’ tendencies towards promoting biases in publication by providing them randomized scenarios. I think the paper itself would make for an interesting discussion in a journal club or a class on the reproducibility crisis.
That said, I’m curious about your response to this paper. Your comment regarding your priors is a bit vague, so I might be misinterpreting you point. Do you mean to say that you think pre-registration of replication attempts would not be valuable because there might be weak biases (as suggested by the paper in question) towards publication of replications that contradict earlier findings? Are you worried about pre-registration or are you worried about replication, or maybe both?
By the way, my take from the paper was that by far the biggest source of bias came from reluctance to publish results that did not meet a significance threshold. We already know about this bias, and have reason to think that in some contexts it’s a really serious obstacle to building robust empirical understanding. Slight biases in publishing ‘gotcha’ replications are less worrisome than these really big publication biases. (by the way, there are interesting ideas for reducing biases in original papers and replications, but reducing both ‘gotcha’ bias and bias against ‘non-significant’ results in publishing direct replications would be easiest)
By “confirms my priors” I just meant that I’m not surprised that there’s publication bias in favor of statistically-significant results (as you say, we knew that already), and that there’s now also publication bias in favor of results that overturn previous results. One effect of replication efforts like Many Labs in psychology is to demonstrate to lots of people that you can get a high-profile paper out of overturning previous results.
I agree with your comments, and don’t think the existence of those publication biases much affects the case for doing replication work.