Friday links: an admirable retraction in ecology, AI vs. mathematics, and more

Also this week: the latest on a major case of serial scientific misconduct in botany, “you cannot be serious,” and more.

From Jeremy:

Theoretical ecologist Priyanga Amarasekare has set up a public website, where she details her responses to UCLA’s latest charges against her. Yes, these latest charges are on top of all the severe sanctions to which she’s already been subjected. That’s perhaps the most dismaying part for me. Not only have the sanctions against her have not only been much more severe than recommended by the departmental committee that investigated the original charges, but UCLA is still finding reasons to continue imposing additional severe sanctions. Also, a reminder of the GoFundMe that Sally Otto set up to help Priyanga support her family and cover her legal expenses. Note that I pass this news on because it is likely to be of interest to some of you, not for purposes of inviting commentary. Previous comment threads on this case, here and elsewhere, have (i) expressed a wide range of views, and (ii) quickly devolved into shouting matches, or threatened to do so. I suspect that most people interested in this case have already heard the full range of views and come to their own opinions (justified or otherwise). I don’t think further discussion here wouldn’t add much value for anyone, and this blog is not going to play host to a shouting match. Comments remain open on this post for purposes of commenting on the other links.

Many of you will recall that, a couple of years ago, whistleblower Ken Thompson and the investigative journalists at Science uncovered serious, repeated data fabrication by Steven Newmaster, a prominent botany prof at Guelph University here in Canada. Guelph “investigated,” by which I mean “cleared Newmaster of misconduct without actually doing any substantial investigation, in the hopes that everyone would forget about the whole thing.” Nice try, Guelph. Ken Thompson and seven other accusers (including several profs from Newmaster’s own department) appealed to the Canadian federal agency responsible for research misconduct. This week, Science reports that the accusers won their appeal. The feds found that the initial investigation was inadequate and ordered Guelph to do it again. The new investigation just found Newmaster guilty of data fabrication in three papers. Guelph remains mum on what sanctions Newmaster will face, if any. But come on–they’ve got to fire him now, surely? I mean, Newmaster’s surely not going to be able to bring in big grants anymore (right?) So even if Guelph were totally unprincipled, what would they gain by retaining him? Anyway, whatever happens to Newmaster, kudos to Ken Thompson and the other accusers for seeing this through. I’m just sorry they had to.

Relatedly: Stuart Buck argues that the swiftness with which scientific misconduct is investigated, and the severity of the sanctions, tends to vary with the nature of the misconduct. At least, with respect to federally funded biomedical research in the US. Spend your grant money on, say, strip clubs? Expect to be fired quickly, and then sent to prison. Spend your grant money on fake research? Expect to keep your job for years (at least), and not be sent to prison. I presume that’s because even the most clear-cut cases of data and image fabrication aren’t as clear-cut as spending grant money on, say, strip clubs? There are various ways to try to explain away deliberate data and image fabrication, and refuting them generally requires at least a bit of expertise. But there’s really no explaining away spending your grant money at a strip club.

Adam Mastroianni channels his inner John McEnroe and argues that you can be serious. Meaning “actually caring about stuff, and for the right reasons.” Great piece. Sample quote, to encourage you to click through:

Before that, I believed what everybody else seems to believe: if you play the game well enough and long enough, eventually you get to stop playing and go do whatever you want. I played the game pretty well for a long time, and now it’s obvious to me that the reward for playing the game is more game. You just keep unlocking levels forever, and the levels don’t even get more interesting (“Ooh, this one is in space!”). It’s just the same thing over and over until you die. You don’t get out by winning; you get out by stopping.

Fields Medalist Terence Tao on how AI will change mathematics research and education. Very interesting.

Questions and answers about applied statistics and multilevel modeling. From statistician Andrew Gelman’s students and Gelman himself.

New satirical campus novel dropped a few months ago, but I only just heard about it. Here’s a review in the Guardian. I’m a huge fan of Small World, and a moderately-sized fan of Moo, so this one’s going on my reading list.

No one is coming to save Canadian higher education over the next decade. The universities are going to have to save themselves, by cutting costs.

Heilmayr et al. (2023 Science) studied the drought sensitivity of mesic forests faced with climate change. The paper was just retracted due to calculation errors. Retraction notice here, Retraction Watch story here. TL;DR: another research group found errors in the R script after the paper was published. Fixing the errors changed the results of some of the robustness checks, and so the authors retracted. Kudos to all involved for doing the right thing and doing it promptly. Mistakes happen in science. People should be praised for correcting them, rather than blamed for committing them in the first place.