Statistics in the news — but not in a good way

(by Andrew Gelman)

Dan Vergano writes in USA Today:

Evidence of plagiarism and complaints about the peer-review process have led a statistics journal to retract a federally funded study that condemned scientific support for global warming. The study, which appeared in 2008 in the journal Computational Statistics and Data Analysis, was headed by statistician Edward Wegman . . . The journal publisher’s legal team “has decided to retract the study,” said CSDA journal editor Stanley Azen . . . A November review by three plagiarism experts of the 2006 congressional report for USA TODAY also concluded that portions contained text from Wikipedia and textbooks. The journal study, co-authored by Wegman student Yasmin Said, detailed part of the congressional report’s analysis.

This sounds bad. Here’s the response:

“Neither Dr. Wegman nor Dr. Said has ever engaged in plagiarism,” says their attorney, Milton Johns, by e-mail. In a March 16 e-mail to the journal, 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. “We would never knowingly publish plagiarized material” wrote Wegman, a former CSDA journal editor.

Wegman’s coauthor on the journal article was his student Yasmin Said, but their attorney said that neither had engaged in plagiarism. So the student they are blaming is someone else.

I did a quick Google scholar on Wegman Said CSDA and found Social networks of author–coauthor relationships, which I assume is the offending article. The authors are Yasmin H. Said, Edward J. Wegman, Wailid K. Sharabati, and John T. Rigsby.

So Wegman and Said are saying that either Sharabati or Rigsby is the plagiarist.

There was also some controversy because the paper was accepted by the journal after five days and apparently not the highest quality, according to CMU expert Kathleen Carley. Here’s what the journal editor wrote, five days after receiving the submission of the ill-fated Said, Wegman, Sharabati, and Rigsby article:

Dear Ed: I personally reviewed your very interesting (and unique) manuscript. I think the paper is very interesting, and I could not identify any errors. So, I am pleased to inform you and your colleagues that your paper “Social Networks of Author-Coauthor Relationships” has been accepted for publication in Computational Statistics and Data Analysis.

Your paper will now be forwarded to the Publisher who will contact you soon with full details. Thank you for submitting your work to this journal.

Vergano writes:

Azen says he must have overseen an earlier, more extensive review of the paper involving outside reviewers. But he says he has no records of this earlier review, because his records were destroyed in an office move. “I would never have done just a personal review,” he says.

As an author and editor of many scientific journals, I actually think it’s ok to just do a personal review. In retrospect, accepting this particular paper was a mistake–but review by a journal editor is a form of peer review. I once submitted a apaper accepted in the same day that I submitted it–and this was to a good journal. I think this is just fine. I don’t see what’s gained by dragging the process out. In this particular case it doesn’t make the journal editor look very good, but that’s a risk you have to take.

I guess what I’m saying is: If Azen really did an earlier review with outside reviewers, then I’m sorry that his records were destroyed because it wouldn’t look so bad to accept such an article if you’re backed up by outside experts. But if actually he actually just reviewed it himself, I think that’s ok as a general policy to occasionally accept excellent papers without wasting everyone’s time in a review process.

There was also another student involved, Denise Reeves; Vergano writes that Reeves was “the ‘most knowledgeable’ person about such analyses on Wegman’s team, according to a note that Wegman sent to CSDA in March.” But Reeves would not seem to be the person accused by Said and Wegman of plagiarism, as she was not a coauthor of the retracted paper. (It would be pretty odd to make a plagiarism accusation against someone who was not a coauthor, as the accusation would imply that a non-author wrote some of the paper, which is a no-no in itself.)

One thing that’s interesting to me is that Said and Wegman are so sure that either Sharabati or Rigsby is the culprit. Perhaps they have some email trail or some manuscript with version control that shows who added the offending material?

Beyond the details, the case is completely bizarre, as noted by John Timmer. In the now-retracted CSDA article, “certain styles of research were suggested to be prone to ‘groupthink, reduced creativity and the possibility of less-rigorous reviewing processes.’ . . . the paper was accepted five days after submission by an editor who was a personal friend of Wegman’s, an irony given that the study was about the nefarious influence of social networks.”

Edward Wegman received the Founders Award in 2002 from the American Statistical Association.

About these ads

60 Responses to “Statistics in the news — but not in a good way”


  1. 1 John May 17, 2011 at 7:46 pm

    Suppose a journal article has 50 authors. Which author is responsible for certifying that none of the text is plagiarized?

  2. 2 The Statistics Forum May 17, 2011 at 8:11 pm

    Hmmm, good question. The whole system seems to operate under the presumption of honesty.

  3. 3 Sebastian May 17, 2011 at 8:14 pm

    John: In the ‘hard’ sciences the first author, with some responsibility falling on the last author who is usually the lab supervisor.

    In the social sciences we don’t have 50 authors and we tend to not order by importance – in those cases every author is expected to have signed of on the paper.

    Andrew: I think there may be instances where lack of review is acceptable – e.g. for a note that corrects an unambiguous error – but I hope we can agree that an article accusing other scientists of misconduct pretty openly should be reviewed, if anything, with extra care.

  4. 4 The Statistics Forum May 17, 2011 at 8:45 pm

    Sebastian:

    Yah, good point. When I sent my letter to the editor of the Journal of Theoretical Biology, I think it got three reviews.

  5. 5 Daniel Lemire May 17, 2011 at 9:55 pm

    Peer review is an honor-based system

    It would take too long to expose all of the flaws of peer review, here are some:

    some work is just flat wrong because the reviewers cannot analyze all of the mathematical results, and because they cannot redo the experiments;
    numerous researchers cheat, sometimes in small ways (“2 out of 3 experiments agree with by theory, let us drop the third one”), sometimes in big ways (“I don’t have time to run these experiments, so let me make up some data”);
    peer review may perpetuate some biases and prevent researchers from putting into question some fundamental questions (“we decided that this is the right way, if you question it, you are a loony”).

    However, for all its faults, peer review remains essential in science. I want other researchers to read and criticize my work. I enjoy it very much when people try to find flaws in my work. I think that my work is serious enough that when people point out flaws, I am usually aware of them at some level and I can respond easily (and enjoy the process).

    The type of peer review I do not enjoy is the country-club approach: 1) does the paper agrees with the goals and views of the reviewers 2) is the paper written by someone we can respect? Fortunately, you can navigate the system and stay away (mostly) from country-club peer review.

    But why do I still like peer review despite its obvious flaws? Because I see it as an honor-based system. In such a system, you have to accept that there will be cheaters. A lot of them. And there will lots of mistakes. All we have to do is be open about it. That is, you cannot say “but my work was peer reviewed so you cannot question it!” or “I am very good, look at these prestigious publications!”. The peer review is there to help the authors. It is not, however, an insurance against fraud or mistakes. I like peer review because it helps me become better, but I do not use the system to determine how good someone else is.

    So, what do we do if we want to know how good someone is? You read his work. You reproduce his experiments. You redo his math. Of course, this scales poorly. If you have to hire someone, you cannot read the work of 50 or 500 candidates. So? I think we have to be realistic. It is hard to know how good someone is. You can get to know 10 or 20 researchers in your life. That is about all.

    Hiring processes are flawed. You will hire cheaters. Get over it.

    Quote from:

    http://lemire.me/blog/archives/2008/08/21/peer-review-is-an-honor-based-system/

  6. 6 The Statistics Forum May 18, 2011 at 12:52 am

    Daniel:

    I don’t think anyone’s planning to hire Wegman–he’s just about at retirement age–but I suppose that his coauthor’s career is now in jeopardy.

    But, just to be clear, in the post above I wasn’t talking about hiring. The issue is that the retracted paper has policy implications.

  7. 7 precip May 18, 2011 at 7:14 am

    It seems that these people have a history of plagiarising which they cannot blame on any students:

    http://deepclimate.org/2011/03/26/wegman-and-said-2011-dubious-scholarship-in-full-colour/

  8. 8 Marco May 18, 2011 at 10:00 am

    Andrew Gelman: the student in question IS Denise Reeves. She has confirmed she was asked to write the section on Social Network Analysis for the so-called Wegman-report. That was heavily plagiarised in itself, but the information so far suggests Reeves was not so much to blame for that (poor instructions from the Wegman-team). Reeves was acknowledged in a rather vague way in the Wegman report.

    For the article itself we thus now have the situation that the four authors appropriated material they did not write themselves, and for which they previously acknowledged, albeit unclearly, Denise Reeves. Clear-cut case of plagiarism, especially since Wegman essentially admits she is the author and that he knew she was, even if we ignore the fact that she plagiarised it from other sources!

    Please note that the blogger who found this plagiarism (Deepclimate.org) has multiple other examples of apparent cut-and-paste from Wegman’s group in other articles. If Wegman wants to blame others for that as well, he must explain how they all managed to converge in his group…

  9. 9 The Statistics Forum May 18, 2011 at 3:23 pm

    Marco:

    Wow–that’s really bizarre that they would have the chutzpah to attribute plagiarism to a student who was not even listed as a coauthor? I mean, who allows a non-coauthor to add huge chunks of material to the paper?

    If true, this really is disgraceful. What’s their rule for coauthorship: if you contribute original material, we’ll list you as a coauthor, but if you plagiarize, you only get “thank you” credit?

    I wonder what the rules are on revoking that Founder’s Award?

  10. 10 omnologos May 18, 2011 at 5:48 pm

    You could go back to 1972 and stop Wegman on his track

  11. 11 Deep Climate May 18, 2011 at 5:58 pm

    Now you’re finally getting it.

    I would also say that Azen’s personal review was totally unacceptable. Does he have any knowledge of the application area? And did he not appreciate the irony of reviewing it himself rather than relying on competent reviewers?

    Also note that his claim that there “must have” been other reviews is highly implausible (notice the lack of definitive assertion), given the five days between submission and Azen’s email to Wegman.

    Oh, and by the way, there’s plenty more “cut-and-paste” evidence to come.

    I hope this will be a big wake-up call for the statistical community. Peer review of applied statistics papers *must* include domain knowledge experts.

    DC

  12. 12 The Statistics Forum May 18, 2011 at 6:18 pm

    Deep:

    I agree that Azen’s review was a mistake in this case and that where there is doubt it is good to get a subject matter expert. I was just saying that I don’t in principle object to a review by the journal editor alone, if that editor is willing to consider himself enough of an expert to make the call. Editorial judgment is a form of peer review if the editor himself is a peer.

    But I’m still reeling at the idea that Said and Wegman blamed the plagiarism on a student who wasn’t even listed as a coauthor. None of the credit, all of the blame, huh?

  13. 13 Anna Haynes May 19, 2011 at 12:18 am

    How typical is it, in peer review, for a resubmitted paper *not* to be shown to the reviewers who critiqued it to verify that their criticisms were adequately addressed – instead, to have the journal editor basically say “hey, here are their criticisms; send it back in saying you’ve addressed them & we’ll publish it”?

  14. 16 Shub Niggurath May 19, 2011 at 3:37 am

    As always, Mr Gelman, there is more to this than meets the eye.

  15. 17 Marco May 19, 2011 at 4:55 am

    Andrew, I think that per the Vancouver Protocol, Wegman and co may rightfully argue she should not be a co-author. But that’s how far I can go with my apologetics…

  16. 18 The Statistics Forum May 19, 2011 at 2:56 pm

    Anna:

    From the correspondence above, it doesn’t look like there were any original reviewers. If there had been, I’d expect Wegman’s resumbission email to have said something like, “We have respondent to the reviewer comments as follows…”, rather than the actual email, which (according to the USA Today story) read:

    Dear Stan: Yasmin Said and I along with student colleagues are submitting a manuscript entitled ―Social Network Analysis of Author-Coauthor Relationships.This was motivated in part by our experience with Congressional Testimony last summer. We introduce the idea of allegiance as a way of clustering these networks. We apply these methods to the coauthor social networks of some prominent scholars and distinguish several fundamentally different modes of co-authorship relations. We also speculate on how these might affect peer review.

    We think this is an interesting and provocative paper. We hope you like it.

    Cheers, Ed Wegman

    Marco:

    I agree that someone’s contributions are 100% plagiarized, then she should not be a coauthor! On the other hand, really the plagiarized parts shouldn’t be in the article in the first place. I don’t see how it can make any sense for the an article to include chunks that aren’t written by the listed authors, without citing the original source.

    • 19 Marco May 19, 2011 at 3:48 pm

      Andrew, I hope my comment was not misunderstood: as per the Vancouver Protocol, even Reeves’ writing an entirely original introduction would not have qualified her for a co-authorship.

      Having said that, not even acknowledging her is at absolute minimum questionable. If I were on a committee having to judge this case, I would not have any problem to rank this (non-)action by itself as plagiarism.

  17. 20 Michael Lavine May 19, 2011 at 3:33 pm

    Marco: The Vancouver Protocol applies to Biomedical journals. For statistics, I think it more appropriate to use the ethical guidelines of the ASA, which are at http://www.amstat.org/committees/ethics/index.html, and which say, in part,

    “Maintain personal responsibility for all work bearing your name; avoid undertaking work or coauthoring publications for which you would not want to acknowledge responsibility. Conversely, accept (or insist upon) appropriate authorship or acknowledgment for professional statistical contributions to research and the resulting publications or testimony. …

    “… identify who is responsible for the statistical work if it would not otherwise be apparent.”

  18. 21 John Mashey May 20, 2011 at 11:00 pm

    I don’t think it’s statistics that’s in the news in a bad way, I think it’s that like any profession, there can be a few bad apples. Given Wegman’s earlier track record, this is especially sad.

    As I wrote in the PDF Deep Climate kindly hosted for me:

    http://deepclimate.files.wordpress.com/2010/09/strange-scholarship-v1-02.pd,

    “All this is strange. I do not think most statisticians try to lie with statistics.”

    Prof. Gelman: you should have gotten some email from me with more information, busy since university spam filters sometimes ar over-enthusiastic, if not, let me know. This will save you some time speculating on information that is actually known.

  19. 22 steven mosher May 20, 2011 at 11:13 pm

    The offending section in the Said paper is section 1.

    The initial portion of section 1 is lifted directly from the co-author Sharabati’s dissertation. It’s highly likely that Sharabati is the sole author of that section.

    Sharabati in short lifted material from his own dissertation ( from the first 31 pages ). At the close of those 31 pages Sharabati says
    “what follows is my original work” In short the first 31 pages are just boiler plate, introductory material.

    So Sharabati as a co author wrote section 1, cribbing from the prefatory matter in his 230 page dissertation.

    Some of this dissertation material, In turn, was cribbed from the wegman report.

    That section of the wegman report was supplied by Reeves.

    Now in reverse:

    Reeves prepared a introduction to SNA for the wegman report as a graduate student. She did not supply citations or footnotes. I guess she thought it was not necessary. The writers of the wegman report took that material into their report and did not look for or catch the lack of proper citation ( whats proper for a report to congress) Parts of the wegman report become fodder for the introduction to Sharabatis dissertation, and then Sharabati uses his own dissertation material to write section 1 of the Said paper. the editors of that journal also did not catch the issue.

    I doubt they would look for plagairism in the throat clearing prefatory material of the paper. Not that they took the time to.

    Ironic. One of Said’s notions is that Pal review corrupts peer review. Of course a Wegman pal reviewed it, missed the plagiarism, and proved the point of the paper.

  20. 23 Katharine May 21, 2011 at 3:18 pm

    I live not far from George Mason.

    The economics department there being a bastion of Austrian economics, complete with its own shrine to Hayek, I am not surprised.

  21. 24 Ted Kirkpatrick May 21, 2011 at 3:45 pm

    Andrew, the controversy over the paper becomes clearer when you read the article in question. If you could take a few minutes to read it (it’s short), I’d like to hear your opinion. Based on your long experience reviewing statistical papers, do you think this one should have been published in its current form? Does the paper seem straightforward or … odd?

    The plagiarism is just a “tell” for bigger problems.

  22. 25 Eli Rabett May 22, 2011 at 12:58 am

    FWIW, social network analysis was never one of Wegman’s strengths, actually he is/was pretty ignorant about it, and Reeves was the member of his group that knew the most (she had taken a one week seminar). That’s what makes the whole thing totally bizarre. The editor who knows nothing about social network analysis accepts a paper from a group that has no track record in social network analysis without outside reviews. C’mon

  23. 27 The Statistics Forum May 22, 2011 at 1:12 pm

    John:

    I’ll explore this further in a future blog, but, just briefly: I didn’t see evidence that Wegman and Said were trying to lie with statistics. I just think it’s less effort to ask a stdent to cut and paste than to go to the trouble of actually becoming an expert in the field of social networks.

    Ted:

    I would not have accepted the paper for CSDA. That said, I’ve been on the editorial boards of many journals, and I’m sure I’ve made lots of mistakes in my editorial decisions.

  24. 28 Gneiss May 22, 2011 at 8:29 pm

    Have you read Deep Climate’s reverse-engineering of statistical analysis presented in the Wegman report? It would be valuable to have more statisticians engaged on the discussion.

    “The Wegman report sees red (noise)”

    http://deepclimate.org/2010/10/25/the-wegman-report-sees-red-noise/

  25. 29 Ted Kirkpatrick May 22, 2011 at 10:11 pm

    Thanks for your assessment, Andrew. I agree that editors occasionally make mistakes, even for excellent journals. Of course, the likelihood of mistakes is much lower when you have multiple reviewers, at least one of which is a specialist in the domain topic (social network analysis in this case). This article apparently only had one reviewer, with no expertise in the domain. The limited review effort highlights the oddity of how this article was accepted.

  26. 30 EarlyCareerStatistician May 23, 2011 at 12:24 am

    There is a major problem where editors accept articles only after so-called ‘personal-review’: from a outsider’s perspective, it’s indistinguishable from nepotism.

    Editors who practice this roll the dice with their credibility…

  27. 31 steven mosher May 23, 2011 at 1:09 am

    Can anyone list what the major conclusions of the paper are and then identify the methodological flaws?

    Bueler? Bueler?

    • 32 The Statistics Forum May 23, 2011 at 2:15 am

      Steven:

      You can find the paper easily enough using Google. The major conclusions are that there are different styles of research collaboration; the methodological flaws are that the entire data analysis is based on four snippets of the collaboration network. There’s no evidence or even argument that you can generalize from these four graphs to the general population, nor is there any evidence or justification of their normative recommendations. The trouble is that the authors didn’t seem to know what they are doing; one piece of evidence of this is that they plagiarized part of the their paper.

  28. 34 dhogaza May 23, 2011 at 10:03 am

    “You can find the paper easily enough using Google”

    Steven “Piltdown Mann” Mosher’s very familiar with the paper and the previous Wegman Report, and is trying to divert the discussion from plagiarism to the fact that this paper doesn’t address the “Mann hockey stick” analysis that formed the meat of the Wegman Report to Congress, only the “social network analysis” part.

    The denialsphere is proclaiming that the plagiarism is unimportant (other than perhaps to Wegman’s reputation and possibly career).

    There are other problems with that analysis comparing Mann’s work with McIntyre’s critique. Wegman was supposed to do an independent review. As DeepClimate has shown, McIntyre provided Wegman with his code and Wegman’s group just re-ran it and presented the results, and McIntyre’s code does some fishy things (mines for the desired result).

    DeepClimate appears to have that nailed, as well. Perhaps the materials e-msiled you by Mashey contains details, if not, I’m sure DeepClimate can quickly point you to his analysis of this aspect of the Wegman Report.

    If you’re interested in following Mosher’s attempted diversion. The other possibility is to say “thanks but no thanks, I’m not playing”.

  29. 35 Gneiss May 23, 2011 at 10:59 am

    A similar criticism, that the broad opinions forming the paper’s conclusions do not follow from its limited, under-described data, was expressed by Kathleen Carley of Carnegie Mellon, the social networks analysis expert consulted by journalist Dan Vergano for his USA Today story.

    http://content.usatoday.com/communities/sciencefair/post/2011/05/retracted-climate-critics-study-panned-by-expert-/1

    “Q: Would you have recommended publication of this paper if you were asked to review it for regular publication — not as an opinion piece — in a standard peer-reviewed network analysis journal?

    A: No – I would have given it a major revision needed.

    Q: (How would you assess the data in this study?)

    A: Data: Compared to many journal articles in the network area the description of the data is quite poor. That is the way the data was collected, the total number of papers, the time span, the method used for selecting articles and so on is not well described.

    Q: (So is what is said in the study wrong?)

    A: Is what is said wrong? As an opinion piece – not really. Is what is said new results? Not really. Perhaps the main “novelty claim” are the definitions of the 4 co-authorship styles. But they haven’t shown what fraction of the data these four account for.”

  30. 36 Gneiss May 23, 2011 at 11:27 am

    The Said, Wegman et al. 2008 paper repeats, admittedly without evidence, the original Wegman report’s accusations that peer review is untrustworthy in paleoclimate research:

    “Indeed, the paleoclimate discussion in Wegman et al. (2006), while showing no hard evidence, does suggest that the papers were refereed with a positive, less-than-critical bias.” (p.2183)

    In contrast, and also without evidence, it asserts the superiority of peer review under the “laboratory” and “mentor” styles of coauthorship favored by Wegman’s group.

    The paper cites only four references, one being the Wegman 2006 report. Repeating that report’s still unsupported allegation against paleoclimatologists stands out as the main point of the Said, Wegman 2008 paper. It does not require expertise in social networks analysis to see this.

    • 37 Lotharsson May 24, 2011 at 12:49 pm

      Despite Mosher’s assertion that this paper’s “review” – apparently undertaken by someone unqualified to review it – somehow “proves the point of the paper”, it seems the counter-argument is far stronger as it:

      a) demonstrates that “laboratory” or “mentor” styles don’t preclude poor review;
      b) suggests that having someone who knows the field review it is important;
      c) demonstrates that post-publication peer review is as important, perhaps even more important, than pre-publication peer review.

  31. 38 omnologos May 23, 2011 at 11:34 am

    I was wondering who would be the first supremely intelligent commenter trying to move away from civilized discourse. It’s dhogaza this time around and that doesn’t surprise me at all.

  32. 39 Michael Chernick May 23, 2011 at 2:48 pm

    I know both Stan Azen and Ed Wegman and I think that some of the accusations are very unfair. It is not uncommon for an editor to know an author. I really strongly doubt that the paper was accepted and accepted quickly because the author and editor are “friends”. Both men are highly respected statisticians. As an author,referee and associate editor with statistics journals I know that the primary purpose of referring papers is to check the accuracy of the statistical content and the originality of the article. Plagarism might be difficult to uncover and checking for it is not typically part of the review process. This should be looked at as a rare and unfortunate error by the author and the editor rather than seen as something evil or nefarious.

    • 40 Lee May 23, 2011 at 11:22 pm

      @Michael Chernick,

      I agree that editor review is not uncommon, but it should only happen when the editor is competent in the subject matter of the paper. In the area of Social Network Analysis, Azen is not – he should have sent that paper out for review. Even more damning for Azen, IMO, is his attempt to cover it up. The record clearly shows that Azen accepted the paper 5 days after submission, and that he did so, in his own words based on his own review.

      “I personally reviewed your very interesting (and unique) manuscript. I think the paper is very interesting, and I could not identify any errors. So, I am pleased to inform you and your colleagues that your paper “Social Networks of Author-Coauthor Relationships” has been accepted for publication…”

      But now Azen is attempting to claim that must have sent it out for independent review, because he never would have accepted it on his own, but he can’t remember to whom, and he has no record of it. Gee

      I suspect that Azen trusted Wegman, for whatever reason, and got burned. He should own up to whatever happened.

      “This should be looked at as a rare and unfortunate error by the author”

      Except that it is not rare, not for Wegman. The blog Deep Climate has documented multiple cases of serial plagiarism in work by Wegman and members of his group. They have also documented that in Wegman’s report to congress, he presented unattributed work from others and claimed it as his own independent work. Worse, the work he took was part of the body of work that he was charged with independently investigating, and that he claimed he had independently investigated.

      This was all laid out in a formal complaint to George Mason University many months ago GMU seems to be dragging their feet on an investigation. One wonders why.

  33. 41 Mark Shapiro May 23, 2011 at 8:38 pm

    Michael,

    While one unfortunate error by Wegman may not be evil, perhaps plagiarism, and replacing analysis with opinion in the service of denigrating the entire paleoclimate community may be.

    Wegman et al cast aspersions on Michael Mann and other paleoclimateologists in multiple publications, plagiarizing multiple sources, ignoring critiques of their work, and failing to analyze the data.

    John,

    The link to your analysis of the Wegman report is broken (missing the “f” at the end). Try:

    http://deepclimate.files.wordpress.com/2010/09/strange-scholarship-v1-02.pdf

  34. 42 Gneiss May 23, 2011 at 10:52 pm

    An article in USA Today last fall (11/21/2010) investigated allegations of plagiarism in the original Wegman et al. 2006 report to congress (which the retracted 2008 paper copied in turn). The USA Today article leads with:
    “An influential 2006 congressional report that raised questions about the validity of global warming research was partly based on material copied from textbooks, Wikipedia and the writings of one of the scientists criticized in the report, plagiarism experts say.”

    Three plagiarism experts are quoted:

    “‘Actually fairly shocking,’ says Cornell physicist Paul Ginsparg by e-mail. ‘My own preliminary appraisal would be guilty as charged.’”

    “‘If I was a peer reviewer of this report and I was to observe the paragraphs they have taken, then I would be obligated to report them,’ says [Skip] Garner of Virginia Tech, who heads a copying detection effort. ‘There are a lot of things in the report that rise to the level of inappropriate.’”

    “‘The plagiarism is fairly obvious when you compare things side-by-side,’ says Ohio State’s Robert Coleman, who chairs OSU’s misconduct committee.”

    Among the sources allegedly plagiarized is Raymond Bradley, who was one of the Wegman report’s targets.

    “But in March, climate scientist Raymond Bradley of the University of Massachusetts asked GMU, based in Fairfax, Va., to investigate ‘clear plagiarism’ of one of his textbooks.”

    “‘The matter is under investigation,’ says GMU spokesman Dan Walsch by e-mail. In a phone interview, Wegman said he could not comment at the university’s request. In an earlier e-mail Wegman sent to Joseph Kunc of the University of Southern California, however, he called the plagiarism charges ‘wild conclusions that have nothing to do with reality.’”

    http://www.usatoday.com/weather/climate/globalwarming/2010-11-21-climate-report-questioned_N.htm

  35. 44 John Mashey May 23, 2011 at 11:55 pm

    Mark: thanks for fix.

    There is, sad to say, a substantial history accumulated, which will become available very soon, and it seems inexplicable to people who knew Wegman a decade or more ago.

    Actually, maybe people her can help me out, if they attended JSM2010.
    Here is a short appendix to be added to SSWR, showing Wegman’s talk, and 2 sessions organized by Said, one of which was very weird. Both were canceled at the last minute.

    Did anyone attend that, hear Wegman speak, and can summarize what he said?

  36. 46 Mark Shapiro May 24, 2011 at 11:47 pm

    “full HUAC?”

    Please.

    John Mashey and DeepClimate together have exactly zero subpoena authority, and they are up against the best politicians, lobbyists, and public relations campaigns that oil money can buy.

    Meanwhile Wegman and Said continue to spread their flawed analyses, based in part on plagiarized material, while GMU seems to stonewall its investigation.

    I bear no animus toward Wegman, and I assume the same for Mashey. But I would like Wegman to tell us all how the report came about, what instructions he had from the investigative committee, and why he only cooperated with McIntyre and not Mann.

    No one has to believe in AGW, but when part of my government defames a group of scientists, I’d like to know why and how.

    You are allowed to disagree.

  37. 47 John Mashey May 25, 2011 at 5:28 am

    So, the dust has settled a bit, given Strange Tales and Emails (STE), a 17-page PDF that annotates Wegman’s email to Elsevier trying to avoid retraction, and other information. People might read that, but academics, especially, should put coffee down first. Keyboards will thank you.

    So, some questions in this thread can now be answered easier.

    1) Sharabati put it in both Said, et al(2008) and his PhD, but the former was accepted July 2007 and the latter October 2008, 14 months later. Somebody fixed “statues” in Said, et al(2008), but no in Sharabati(2008) or the later Rezazad(21009), who Wegman does not mention.

    Wegman passing (plagiarized) material to students to incorporate elsewhere is not uncommon. in STE Appendix B.1:
    a-b seems to have worked that way.
    d-f-g-h-i certainly did.
    m-n, m-o, and maybe m-p did also.

    2) STE certainly hints at how Kathleen Carley got asked by Vergano – she was mentioned as an expert in Wegman’s email. As noted, there are at least 2 other SNA researchers who’ve looked at this and say similar things. One of them (the unnamed one) had this wry comment last week:
    “Too bad you can only retract papers when it turns out they were plagiarized, when they should be retracted for not having any coherent or sensible argument!”

    3) For Michael Chernick:
    Can you say more about which accusations are unfair?
    As I wrote in STE, p.2:

    “The, second, more unusual problem was an obvious breach of peer review process,6
    6 Plagiarism can be easy to miss, but the 2nd problem is bypassing credible peer review and thus missing other serious issues SNA experts see quickly.”

    I would not expect a reviewer to notice plagiarism, especially from a field with which they are unfamiliar … but then of course, they shouldn’t be reviewing it alone. I wouldn’t necessarily expect someone familiar with the field to see it either, as experts tend to skip introductory material. Finally, with a well-known senior professor’s name on a paper, it’s just not something most would think of.

    (Bradley missed the tree-rings plagiarism in the Wegman Report, much to his chagrin later.) Still, a page-and-half of text, with one citation from 1973 (see SSWR, p.118, is odd.

    The Granovetter(1973) reference was clearly inserted after the WR.
    it is vaguely relevant, but a bit odd to insert there. It is of course a vastly-cited paper, but it “feels” like a false citation inserted by people who don’t know what they are doing, since the Said(2008) never uses the idea of weak ties. Obviously it is impossible to prove they never read the paper. However, much of the introduction just isn’t relevant to the rest of the paper.

    On a somewhat different tack, section 2 of Said(2008) talks about “allegiance”, a clustering method developed by Rigsby. As per SSWR p.145, it is completely unclear whether this is a new idea or just a new name invented for a technique created long ago. Again, Wegman thought Reese the most SNA-knowledgeable in the group.

  38. 51 Eli Rabett May 25, 2011 at 11:11 am

    In addition to the sin of self plagiarism (aka recycling comments at different blogs without footnotes), perhaps omno would like to accuse Eli of self promotion, it’s not so much Wegman’s serial plagiarism as his reaction to it when pointed out.

    Of course, the statistical analysis in the Wegman report and elsewhere, are, shall we say, interesting. In this regard here is the right place to paraphrase Gerald North’s reply to Edward Wegman’s assertion that statistician should be involved in paleoclimate work: For the most part they are not interested.

  39. 52 omnologos May 25, 2011 at 2:58 pm

    The HUAC bit refers to the practice of reporting as somehow relevant the most tiny of minutiae like a session getting cancelled at a conference. I doubt many people of whatever repute would survive such an onslaught unscathed.

  40. 53 Eli Rabett May 25, 2011 at 4:40 pm

    Well, perhaps not the place to deal with this, but Steve McIntyre, whose work Wegman used to try and discredit Michael Mann, has been moaning for seven or eight years as somehow relevant the most tiny of minutiae like how Mann lied when he said he sent McIntyre and Excel file when it was a txt file that was obviously created in Excel.

    Omno, would that Eli could say this brings to mind Bismark’s take on the Schleswig-Holstein problem, that there are three people who understood it. One is dead, the other gone crazy, and Eli, Eli has forgotten all about it.

    • 54 omnologos May 25, 2011 at 5:35 pm

      even if I find Eli’s style a reminder of something gone amiss at the top floors let me state that for McIntyre to go HUAC he would have to start writing about the most obscure of Mann’s students . Has he so far?

    • 55 Steve McIntyre May 26, 2011 at 4:43 pm

      As so often, Eli Rabett is spreading disinformation. I asked Mann for the FTP site of the data used in MBH98. He said that he had “forgotten” the location and said that Scott Rutherford would locate the data for me. Rutherford then referred me to a URL at the U of Virginia website.

      I sent Mann the data set that I downloaded and asked him to confirm that it was the correct data. He said that no one else had had trouble with it.

      The data set had various problems. Mann falsely claimed that I had asked for the data in an Excel spreadsheet and that errors had been introduced in the preparation of the spreadsheet. This was untrue on at least two counts. I hadn’t asked for an Excel spreadsheet. My interest was in the data as used. Not only did I not ask for an Excel spreadsheet, if offered, I would have said that I wanted the data in the form that it was used. Secondly, the data set that I downloaded was dated nearly a year prior to my request to Mann and was therefore not prepared in response to my request.

      The Climategate emails show that even CRU were wary of endorsing Mann’s claims about Excel spreadsheets, pointing out that I would simply show my email correspondence, which I did.

      The Climategate emails also show that Mann provided CRU in summer 2003 with data that he refused to provide to me. In that email, he told CRU that the data was his “dirty laundry” and was provided to CRU only because they were “trusted” colleagues.

      All of this information has been known for a long time.

  41. 56 Mark Shapiro May 25, 2011 at 9:08 pm

    Clean energy will make us healthier, wealthier, safer, and more secure.

    Why do so many (otherwise) good folks insist on dirty energy?

  42. 58 Eli Rabett May 26, 2011 at 3:32 am

    Yes, see Scott Rutherford among others. Rutherford was the poor grad student Mann asked to get the data together to send to McIntyre. He made some obvious copying errors (files were displaced by a year, etc.) and was jumped on by McI

    FWIW, Said is Wegman’s pride and joy, not the most obscure of his students. Really Om this entire thing is so strange that you really don’t want to jump in.

  43. 59 John Mashey May 26, 2011 at 11:49 pm

    Regarding personal reviews:

    Azen told Vergano:
    “I would never have done just a personal review,”

    See p.149 of SSWR.

    In the issue in which this appeared,of the 31 articles graphed, one was 0 days, one was 1 day, said(2008) was 6 days, and everything else was 61+ days.

    So it looks like there were 3 with personal reviews, but the other two were discussing resampling techniques (bootstrap, jackknife) on which Azen has often written.


  1. 1 More statistical hijinks on climate change | Sinting Link Trackback on May 21, 2011 at 3:07 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




About

The Statistics Forum, brought to you by the American Statistical Association and CHANCE magazine, provides everyone the opportunity to participate in discussions about probability and statistics and their role in important and interesting topics.

The views expressed here are those of the individual authors and not necessarily those of the ASA, its officers, or its staff. The Statistics Forum is edited by Andrew Gelman.

A Magazine for People Interested in the Analysis of Data

Follow

Get every new post delivered to your Inbox.

Join 57 other followers

%d bloggers like this: