Join 3,523 readers in helping fund MetaFilter (Hide)


More statistical hijinks on climate change
May 21, 2011 7:56 AM   Subscribe

“certain styles of research were suggested to be prone to ‘groupthink, reduced creativity and the possibility of less-rigorous reviewing processes.’ Edward Wegman is a professor at George Mason and a distinguished statistician with a long career, a former winner of the ASA's Founders Award. In 2006 he testified before Congress on climate science, sharply criticizing the statistical methodology of Michael Mann's "hockey stick graph," which showed a sharp increase in global temperature in the last part of the 20th century. One section of Wegman's testimony concerned "social network analysis," and suggested that Mann's tightly-knit network of co-authors might have led to insufficiently aggressive peer review. USA Today reports that Wegman's testimony contained a substantial quantity of plagiarized material, and the peer-reviewed article derived from the testimony has been retracted by the journal that published it. John Mashey has compiled an obsessively thorough catalogue of the plagiarized text. (large .pdf.)

More USA Today reporting: The journal editor, a friend of Wegman's, accepted the paper 5 days after submission. He insists he got a referee report from an outside reviewer, but that all records of this report were lost when he moved offices. Network analysis expert Kathleen Carley at CMU, asked to read the paper, said, "I see the paper as more of an opinion piece... No data is provided to support this argument."
posted by escabeche (26 comments total) 19 users marked this as a favorite

 
This is my surprised fa-- ah, shit. Never mind. *sigh*
posted by verb at 8:10 AM on May 21, 2011 [2 favorites]


The most surprising thing about this post is that USA Today appears to be doing real investigative journalism.
posted by craichead at 8:13 AM on May 21, 2011 [11 favorites]



The most surprising thing about this post is that USA Today appears to be doing real investigative journalism.


I believe this statement was made word for word in the last USA Today linked FPP.
posted by spicynuts at 8:18 AM on May 21, 2011 [5 favorites]


The question now is: How long is he going to be hung by his thumbs in Times Square?
posted by dunkadunc at 8:20 AM on May 21, 2011 [2 favorites]


The real question is: Who was paying him to do it?
posted by DU at 8:21 AM on May 21, 2011 [5 favorites]


Things I am not surprised about: Smokey Joe Barton requested the congressional report involved. If you want to follow the money, Smokey Joe is a wholly-owned subsidiary of the energy industry. I don't know who funded the report, but who funds him is never in doubt.
posted by immlass at 8:24 AM on May 21, 2011


after 7 years as a graduate student in a top research institution, plagiarism in academia no longer surprises me. there are huge egos in all strata and very little open discussion or alarm at this state of affairs.
posted by pistachio at 8:24 AM on May 21, 2011 [5 favorites]


The most surprising thing about this post is that USA Today appears to be doing real investigative journalism.

Perhaps they just noticed a vacuum in the market.
posted by brennen at 8:28 AM on May 21, 2011 [3 favorites]


I believe this statement was made word for word in the last USA Today linked FPP.

Then show us the data. What have you got to hide?
posted by notyou at 8:28 AM on May 21, 2011


Plagiarism experts told USA Today that the report, which was authored by George Mason University statistician Edward Wegman, copied material from textbooks, Wikipedia, and other sources.

John Mashey, a retired computer scientist, conducted a year-long analysis of the Wegman report and found that 35 of the report's 91 pages "are mostly plagiarized text, but often injected with errors, bias and changes of meaning."


Copying Wikipedia ain't just for dumb high school kids anymore.
posted by 2N2222 at 8:36 AM on May 21, 2011


GMU? Ya don't say.
posted by symbioid at 8:38 AM on May 21, 2011 [2 favorites]


Well, now it's three times in five weeks.
posted by absalom at 8:48 AM on May 21, 2011 [4 favorites]


What the fuck kind of professor plagarizes from Wikipedia? I'm a piddling undergrad at a no-name university in the southwest and I'm fucking smarter than that. This is insane.

Nobody that stupid should be a professor.
posted by Avenger at 8:49 AM on May 21, 2011


And it's so easy to write something in your own words, too. Copying and pasting is just... so lazy.
posted by dunkadunc at 8:50 AM on May 21, 2011


I doubt he really plagiarized from Wikipedia: probably he plagiarized from the same source the Wikipedia text was stolen from. This is a common result when I do plagiarism analyses of student papers. Still....
posted by anotherpanacea at 9:36 AM on May 21, 2011


The real question is: Who was paying him to do it?
posted by DU at 8:21 AM


It seems very likely to me he was paid, and almost certain that he was recruited for the job and handed text prepared in advance that he did little more to than read over for the sake of a smooth delivery.

Unless he's deep into his decline, I can't see a person with his reputation delivering heavily plagiarized testimony before Congress. He certainly must have known it would be closely scrutinized.
posted by jamjam at 10:39 AM on May 21, 2011


Virginia resisted should write their AG and demand he investigate this illegal use of state money for fraud, just as he's been investigating Mann.
posted by humanfont at 10:51 AM on May 21, 2011 [2 favorites]


Let me join the copy-and-paste bandwagon, by C&Ping something from Crooked Timber almost 5 years ago:

In other words, if I understand things correctly, the first key finding is that (drumroll) Mann has co-authored a paper with every one of his co-authors. This obviously demonstrates his “centrality” to the group consisting of his co-authors.

Instead of lauding the USA Today for investigative reporting, people should be asking why they're a day (half-decade, but who's counting?) late and a dollar short in calling out some total bullshit that was given credence at the time. How piss-poor is the national US press that this is held up as some sort of example of how things should be done?
posted by Jakey at 12:01 PM on May 21, 2011


> What the fuck kind of professor plagarizes from Wikipedia?
> I doubt he really plagiarized from Wikipedia

Wegman's going with the classic "it was one of my pesky underlings" denial. From the USA Today article:
Wegman blamed a student who "had basically copied and pasted" from others' work into the 2006 congressional report, and said the text was lifted without acknowledgment and used in the journal study.
I find it a little odd that Wegman is admitting to lifting text without acknowledgement, but justifying it because it was text written by one of his students, but I guess that's about the best he can do under the circumstances.

However, in the John R. Mashey report, Mashey says the following:
Much of the writing, perhaps even most, was done by the 3rd author, Wegman student Yasmin H. Said, PhD Spring 2005. (p 3)
Said's 2005 dissertation appears to also contain plagiarised text (p 87), and it's plagiarised in much the same way as a lot of the text of the Wegman Report: loose re-writes that copy long phrases and change the odd word, but are structured identically to the source. However, the Wegman report also contains entire paragraphs that've been copied-and-pasted, which Said's dissertation apparently does not.

Although Dr Wegman is standing with Dr Said at this time, I wouldn't be surprised if she gets thrown under the bus if the pressure keeps building. If she does, it will be to obscure the true issue: that Wegman put his name to a report that deliberately distorts the scientific facts about climate change. Whether he wrote it himself, copied it from Wikipedia or was handed the material by a third party and then passed it off as his own, his behaviour is shameful.
posted by Georgina at 12:10 PM on May 21, 2011


illegal use of state money for fraud, just as he's been investigating Mann.

That should be done, I agree 100%. It could be done by groups like ACLU, Southern Poverty Law Center, etc global warming is more serious in the long run than racism, terrorism and other ills, why is there no legal advocacy group that can sue people who are putting out demonstratively incorrect information.
posted by stbalbach at 12:12 PM on May 21, 2011


The investigative journalism that USA Today did, Jakey, wasn't that they pointed out that the paper sucked. It was that they used the Freedom of Information Act to get evidence that the journal editor published the paper without submitting it for peer review. The problem with pointing out that the paper sucks is that people are going to then ask "well, how did it pass peer review"? And the answer here appears to be that it didn't.
posted by craichead at 12:19 PM on May 21, 2011 [2 favorites]


No it shouldn't be done. Virginia AG Ken Cuccinelli has been investigating Mann with a Grand Jury as an act of global warming denialist showboating to his conservative base.

Out of curiousity does this finding of plagiarism refute the underlying claims, or does this just show that the work was not his own?
posted by humanfont at 12:22 PM on May 21, 2011


The Wikipedia page on the hockey stick "controversy" is actually pretty well fleshed out.
posted by Wyatt at 12:43 PM on May 21, 2011


My two cents: this wasn't even a good paper to begin with...

The introduction is pretty poorly written, despite (or because of) the plagiarism? Incredibly choppy and repetitive in places.

It's amusing that the paper was published in a journal on statistics, as he considers social networks based on an individual Principal Investigator (PI), of size no larger than 101. There are four specific examples of networks, defining four different "styles." This seems like a tiny sample, and probably not statistically significant.

There's also no overlap between the networks of big name authors. All of the conclusions are apparently predicated on the idea that anonymous referees are always drawn from a pool of the PI's previous coauthors, which is almost certainly false.

From the conclusions:

Of course because referees are not identified, getting hard evidence of independence, unbiasedness and knowledgeable expertise is not readily available. The social network analysis can therefore only be suggestive. It is our contention, however, that safeguards such as double blind refereeing and not identifying referees invariably lead to the conclusion that peer review is at best an imperfect system. Anyone with a long history of publication in their heart of hearts knows that they have benefited or have been penalized, probably both, by imperfect peer review.


This is an argument against peer review itself, and is independent of their results. Unless a machine does the reviewing, it's impossible to prevent bias. And, as he points out, there's no evidence for it in the paper.
posted by bessel functions seem unnecessarily complicated at 12:43 PM on May 21, 2011


Isn't GMU where a bunch of the more... interesting... guys at Volokh teach? Is the place a hotbed of right libertarianism?
posted by Justinian at 2:17 PM on May 21, 2011 [1 favorite]


GMU is owned outright by the Koch brothers.
posted by Peztopiary at 3:30 AM on May 22, 2011 [1 favorite]


« Older Filtering out porn algorithmically takes audio int...  |  The story begins in 1879. Chev... Newer »


This thread has been archived and is closed to new comments