Mon: “Edlin’s rule” for routinely scaling down published estimates
Tues: Basketball Stats: Don’t model the probability of win, model the expected score differential
Wed: A good comment on one of my papers
Thurs: “What Can we Learn from the Many Labs Replication Project?”
Fri: God/leaf/tree
Sat: “We are moving from an era of private data and public analyses to one of public data and private analyses. Just as we have learned to be cautious about data that are missing, we may have to be cautious about missing analyses also.”
(Reading some of your old blog posts on Wegman)
I think you are quite right to speak out about plagiarism. And right in being angry at denialism or failure to take responsibility. But Wegman really doesn’t deserve a comparison to Stapel (forging data all over the place). And if he is really a serial plagiarizer (I know the type), you would be seeing it in all his earlier work, not just around the Said association and the 2006 climate study. I just think you are being a bit unfair to compare him to Stapel in all those old posts.
All that said, Weggie should have quickly sent in corrections for background passages, etc. lifted from Wiki and added another site and quotations and a remark in text for the Bradley stuff.
But the Stapel stuff comes off as a little much, dude.
Nony:
The Wegman stuff disturbs me because he’s a statistician, also because he does the irritating thing of denying it even after he got caught. And of course it was particularly uncool to copy from Wikipedia and introduce a math error in the process. But I think my interest in his case came mostly because he’s a prominent statistician. A “Wegman” in another field would not interest me so much.
I agree that he’s different from Stapel. Wegman had a long and successful career which was pretty much based on administration, so plagiarism must have come naturally to him; he didn’t make his name based on original research. If you look him up, you’ll see that his most-cited articles are reviews. In contrast, Stapel was going around claiming to do original research, so plagiarism wouldn’t have worked for him. He had to flat-out make things up.
What’s the allusion to plagiarism coming naturally to administrators? Can you elaborate?
Rahul:
Perhaps “administrator” was not the right word. What I was trying to say that Wegman made his name writing review articles and in various service capacities (hence his service award from the American Statistical Association), tasks for which originality is not always required. Of course, lots of people (me, for example!) write review articles and textbooks without plagiarizing, but by point was that if you’re doing nothing but writing reviews, it doesn’t seem so far to slip into plagiarism. But for original research such as Stapel, you need to go further, in his case via fraud.
Ah, ok. I see.
As an aside, do you think the definition of plagiarism is contextual? In the sense do some kinds of documents / some fields cite more profusely than others? Is this related to a need or a convention?
I agree that he should have immediately corrected the problems. And I understand your annoyance with failure to admit things. About 95% of the Internet debators piss me off on this.
I don’t think he was a typical plagiarizer though. You’re not seeing the bulk of the papers plagiarized. Not seeing it throughout his earlier work. Few sections in background.
And yes…it’s still wrong. Just, it doesn’t have the feel of the typical overall plagiarizer.
In the current scientific establishment is the penalty for a one-off plagiarizer any less harsh than for a serial plagiarizer? Penalty by convention, boycott, ridicule etc.
Perhaps that’s one reason why people loathe admitting it?
Nony:
Yes, my impression (not based on any inside knowledge) was that Wegman’s late-career plagiarisms are products of good intentions: he agreed to write that report for the congressional committee, he volunteered to write review articles for the journal he was editing, he was trying to help out students he was supervising . . . basically, he agreed to do a bunch of stuff that neither he nor his collaborators were able or willing to do. Then instead of admitting he couldn’t do it (or putting in the late nights to actually do the work), he started cheating and covering up. As to his earlier career, I just don’t know. It’s possible he was cutting lots of corners back then too, or maybe not.
He cut some corners on the stats analysis too. He basically copied some of Steve McIntyre’s crits without fully understanding them, without understanding Mann’s work, too. And I’m not accusing him of copying someone’s effort. That was sufficiently referenced. The issue is that he didn’t really work through the particulars but lent his authority to them. Really, despite McI’s failings he’s much more clued in to how to unravel the Mann algorithm and its impacts (and that thing is complex).
And I say this as someone who wanted Mann criticized, just criticized thoughtfully. Jollife (time series expert who wrote books and such, more of an expert on the math than Mann and critical of some choices of Mann) said that to really understand/adjudicate the work would take a lot of work, access to code, and probably ability to question the participants. So there was a certain lack of intellectual humility by Wegman.
I’m looking forward to Friday’s post. Maybe I’ll finally figure out what the response variable is in the model formula Y~God/leaf/tree….Knowledge? Insight? Wisdom? Fame? Fortune? Glory? Serenity? Peace? Progress? Or something else entirely….?
Or maybe it should be Y~tree/leaf/God?