Also this week: nerd sniping Stephen Heard, what’s a better word for p-hacking, the latest modern aRt, worst zoo exhibit ever, and more.
From the Onion: Zoo posts hourly updates on aphid about to give birth. Clearly I need to start posting updates on our Daphnia!
Zombie ideas in conservation biology: 3/4 of studies finding significant effects of habitat fragmentation per se (as opposed to habitat loss) find only positive significant effects–but only 40% of those studies describe their results that way in their abstracts. In contrast, the large majority of studies that find only negative significant effects of fragmentation describe their results that way in their abstracts. Studies finding positive effects of fragmentation also sometimes emphasize that their results shouldn’t be extrapolated to other systems–but studies finding negative effects of fragmentation never emphasize that. Apparently, authors tend to spin their abstracts to line up with what they expected (or hoped?) to find, rather than with what they actually found? It would be very interesting to go through the same exercise for papers on other topics. (ht Mark Vellend, via the comments)
Ecologist Joe Craine was fired from his
tenured position at Kansas State University after the university concluded that his misconduct accusations against two KSU colleagues were “malicious, or at the very least frivolous”. NSF denied his application for federal whistleblower protection because “baseless accusations of fraud to the Ecology editors” are not eligible for federal whistleblower protection. Now the US Circuit Court of Appeals for the 10th Circuit has upheld that denial. Follow the link, and see here, for background on this unusual case (I know nothing besides what I’ve read in news reports). I would not take this case as illustrative or emblematic of any broader problem or issue. Perhaps the only broader lessons I would take away are that, if you want to accuse somebody of scientific misconduct (as opposed to merely wanting to correct published errors, without making any claims about the reasons for those errors), you should (i) have a valid basis for doing so, and (ii) speak to a lawyer and determine the proper channels to go through before making the accusation. (UPDATE: paragraph corrected)
Terry McGlynn on how he handles bad reviews, as both an author and editor. Very useful perspective, includes several points that aren’t made often enough.
Andrew Gelman doesn’t like the term “p-hacking” because it suggests conscious, nefarious intent on the part of the investigator that usually isn’t there. I agree with the suggestion in the comments that “researcher degrees of freedom” is the best term. Will try to remember to stick to that term in future (I’ve used “p-hacking” myself in the past without meaning to imply nefarious intent).
A brief potted history of public intellectuals, how they’ve made a living, and how their means of making a living has affected their work. Useful historical context for present discussions of the public role of scientists. Worth keeping in mind that what we all think of as the “normal” state of affairs–public intellectuals mostly are college and university faculty–only really goes back to WWII.
Economists like to think of themselves as scientists. So here’s an economist asking what’s so great about science. Long, wide-ranging post, covering a lot of philosophy of science and its applicability (or otherwise) to economics. You may want to skim for the interesting nuggets, for instance on how statistical methods of time series analysis struggled to gain a foothold in economics. Always enlightening to read about how different fields have handled the same issues ecologists have struggled with.
Accidental aRt remains very amusing.
I would love for Stephen Heard to write a blog post about the Latin name of one of these species. 😉 (ht Matt Levine)