Also this week: the world’s luckiest ecologist, why you should move to Tulsa, R package vs. researcher degrees of freedom, app vs. LaTeX, when x>x, the Good
Place Class, and more. Lots of good stuff this week!
UPDATE: Of all the links to be late to:
Apparently there’s now an app that can take a screenshot of an equation and automatically convert it to LaTeX. Even if the equation is handwritten. If only it worked with MS Equation Editor too, lamented this non-LaTeX user. (ht @noahpinion)
Queen’s University is running a 2-day summer workshop on modeling structured populations. Looks like they have plenty of good speakers lined up (Cherie Briggs, Ottar Bjørnstad, Aaron King, Helen Wearing…) Follow the link for information on how to attend and present a poster.
An R package that attempts to count and visualize “researcher degrees of freedom”, exhaustively and automatically. Certainly useful as a teaching tool, potentially useful in research as well. Though there are surely aspects of researcher degrees of freedom that resist enumeration, aren’t there? (ht Andrew Gelman)
A statistical profile of Twitter-using economists and how they use Twitter. Interesting. I’d be curious to see a similar analysis for ecology Twitter.
Anna Grear vs. John Quiggin on whether we should extend human rights to apply to nature as a way to protect and conserve nature. I think I disagree with both of them, but I don’t know much about the topic.
Data on what predicts undergraduate major choice in the US. Depending on your temperament, you will be either amused or disturbed to learn the extent to which random happenstance (e.g., the times of day at which classes are scheduled) apparently predicts choice of major. Note that I haven’t read the studies underlying the linked news article, so can’t vouch for them. Possibly, some of the results are p-hacked or type M errors or etc. FWIW (maybe not much), some of the results ring very true to me, such as that biology is among the majors most often dropped. But I’m mostly just passing this along so you can follow up yourself if you’re interested.
An argument that highly educated people and “creatives” should consider moving to mid-sized US cities, even those that are perceived as “uncool” and located in rural and politically-conservative states. (ht @noahpinion) Basically: if you’d consider living and working in (say) Austin, Portland OR, or Asheville, you should consider (say) Tulsa, Des Moines, and Boise too. Because you’ll get similar amenities and “vibe”, only much cheaper. I link to this because I wonder if the same argument applies to some ecology faculty job seekers. Ecology faculty jobs in rural states tend to get relatively few applicants (though still many in an absolute sense). No doubt there are good reasons for that–I’m sure many ecology faculty job seekers who pass on applying for jobs in places like Tulsa and Des Moines really wouldn’t be happy living and working in those places. But is that true for everybody who chooses not to apply to ecology faculty jobs in such places? I dunno, what do you think? Maybe a different way to ask the same question (or a similar question) is: how common do you think it is for ecology faculty job seekers (or anyone really) to move someplace and then be surprised at how much or how little they like it? As a postdoc, I had the experience of moving somewhere I hadn’t ever visited, that was very different from any place I’d ever lived before and very far from family, and being pleasantly surprised how much I liked it. That is, I didn’t know enough about either the place or my own preferences to accurately predict how well I’d like it. Does that make me unusual?
An admirably clear and accessible overview of new work debunking recent high profile attempts to detect geographic variation in selection on genetic variation for human height across Europe. I link to this mainly because it got me thinking back to our old post on “small effects” and when they can matter. Here, very slight biases in effect size estimates end up mattering a lot because they’re consistent biases across many, many genetic loci.
Turns out that huge piles of research associating smartphone and social media use with bad health outcomes are unreliable because they all depend on self-reports of smartphone and social media use. Self-reported use turns out to be only very loosely correlated with actual use. I link to this mainly because it got me thinking back to Meghan’s old post on the 80 hour academic work week myth. People’s self-reports of how much time they spend doing anything are basically rubbish, except maybe for scheduled tasks.
Continuing with the theme of “linking to stuff that got me thinking back to old posts”: there’s no such thing as being on the wrong side of history. Which got me thinking back to this old post asking if there are any currently widely-accepted scientific practices that will someday be seen as unethical.
It’s very hard to apply basic choice theory to Brexit. I leave it to others better-qualified than me to decide if this illustrates something about choice theory, something about Brexit, both, or neither.
Be careful out there Meghan. 🙂 (ht @dandrezner)
This makes me feel like I should order more used books from Amazon: