Friday links: valuing scientists vs. science, real stats vs. fake data, Pigliucci vs. Tyson, and more (UPDATED)

Also this week: the factors influencing student success in college, how (most) authors choose journals, makessense stop!, a pointer to a great comment thread, and much more. Oh, and extreme beer nerdery.

From Meg:

This NY Times Magazine article focuses on the factors that influence success in college, and efforts at the University of Texas-Austin to improve retention and graduation rates, particularly of underrepresented minorities and lower income students. It relates to the series of posts that I had on stereotype threat. ht: Peter Adler

From Jeremy:

Paleontologist John Hutchinson has a lengthy, thoughtful piece on trends in scientific openness, self-correction, and post-publication review. Prompted in part by his own experiences in voluntarily going through the lengthy formal procedures his university has for formally correcting papers published by its employees. And also by his observations of a high-profile case of post-publication review in paleontology that seems to have gone rather suboptimally. Echoes many of my own thoughts. Like John, I think we currently live in interesting times. Clashes about issues like post-publication review often come down to deep-seated and irresolvable disagreements about values. About things as basic as how to value individuals (e.g, “scientists”) vs. collective enterprises far bigger than any one individual (e.g., “science”).

Another very prominent social psychologist seems to have been faking data for years. The evidence against him is purely statistical, but nonetheless damning. Data Colada has a very nice and highly accessible post on the stats. It strikes me as a good example to use in introductory undergraduate statistics classes. The example should grab their interest, and the analysis is sophisticated enough to make them think but not so sophisticated as to be inaccessible. It also strikes me as a good example of a case where rejecting or failing to reject a null hypothesis is of great scientific interest. The goal is to infer, beyond a reasonable doubt, whether or not a particular person has or hasn’t faked data. We don’t want to quantify our personal beliefs about whether he did or not, as a subjective Bayesian would. We don’t even want to quantify the fraction of relevantly-similar cases in which the data were faked, as an objective Bayesian like Nate Silver would. Rather, we have a hypothesis we want to subject to a severe test. People who make the blanket claim that frequentist hypothesis testing is never of scientific interest are just wrong. Further discussion from Deborah Mayo and Neuroskeptic here.

Evolutionary biologist Massimo Pigliucci vs. physicist Neil deGrasse Tyson on whether scientists should read more philosophy. I’m with Massimo on this one, I think all scientists should read some philosophy of science.

Paleoecologist Jacquelyn Gill wants you to tell the story of your post-Ph.D. career path (non-academic or academic) for a blogging carnival. Click through for details on how to participate. And here (UPDATE: link added) is the latest post in our own intermittent-but-ongoing series on non-academic careers for ecologists.

Ever assigned undergraduate students to write an essay in which they argue for or against some statement or position, and gotten back lots of essays with arguments so bad they don’t even qualify as arguments? And maybe found that they didn’t get any better with practice or with feedback from you? Read this, it’ll help. Oh, and there’s a great zinger at the end, but only people who’ve read a fair bit of philosophy will get it.

PEGE Journal Club has a very good discussion of recent work by Estes & Arnold on resolving the “paradox of stasis” in macroevolution. I liked it just because it hits on a number of themes that are near and dear to my heart (and I think to Brian’s as well): the importance (and difficulty) of linking microscale processes and macroscale patterns, model-fitting approaches for (tentatively) inferring process from pattern, the value of avoiding statistical machismo, the bias-variance trade-off, and more.

A snapshot of public higher education funding in the US. Improving recently in most states, but still much worse than before the recession.

Survey data on how Canadian scientists choose where to publish. Apparently, I’m a fairly typical Canadian scientist. Of course, you’re free to make different choices if you want. It’s up to each of us to choose our own path in science.

Zombie ideas, legal edition. 🙂

And finally, a chain of pubs has been tweeting my post on Fisher’s geometrical model vs. IPA. Gives new meaning to the phrase “beer nerd.” 🙂

Hoisted from the comments:

Want to read a really great, high-level discussion of big ideas in tropical ecology and how to test them? With some side discussions on scientific rhetoric and the influence of science blogs thrown in? Look no further! UPDATE: And one of the commenters, Carina Baskett, has continued the conversation with a post of her own.