Retraction of Said, Wegman et al 2008, part 2

Early climate contrarian reactions to the retraction of Said, Wegman et al 2008 have grasped at straws, holding that this does not affect the findings of the paper and the earlier Wegman report alleging inadequate peer review in climate science.

Now USA Today’s Dan Vergano, who broke the the retraction story, addresses exactly that contention in a follow up piece. Social network analysis expert Kathleen Carley of Carnegie Mellon calls Said et al “more of an opinion piece” that would have required “major revision” to render it fit for publication in an SNA journal.

And it gets worse. Computational Statistics and Data Analysis chief editor Stanley Azen “personally reviewed” the paper and sent Wegman an acceptance notice within days of submission.  Meanwhile, Virginia Tech’s Skip Garner enumerates the potential consequences of the research misconduct finding, including the possible need to investigate “ethical issues such as conflict-of-interest, haste vs. scientific rigor and bias”.

Vergano’s devastating piece quotes liberally from his email interview with Carley, which focused on Said et al’s methodology in examining the relative merits of the “entrepreneurial”  and “mentor” styles of co-authorship [updated 2011/05/16]:

The authors speculate that the entrepreneurial style leads to peer review abuse. No data is provided to support this argument. … [Emphasis added]

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.

By the way, it seems there is a connection of Carley to the Wegman team.

[Carley] even taught the one-week course that one of Wegman’s students took before 2006, making the student the “most knowledgeable” person about such analyses on Wegman’s team, according to a note that Wegman sent to CSDA in March.

So this appears to be the unnamed student expert who provided SNA background material used in Said et al and on whom Wegman is clumsily attempting to shift responsibility. On top of everything else, that speaks volumes about the actual credited authors’ supposed expertise. And it does cast a new light on Yasmin Said’s stated areas of expertise back in July 2007 when the paper was submitted and she was still associate editor at the CSDA journal.

Y.H. Said …. Rockville, MD, USA

Biostatistics, Epidemiology, Public Health, Statistical Modeling and Graphics, Adaptive Design, Social Network Theory, Data Mining, Time Series Analysis, Computer Intrusion Detection, Climatology, Metadata [Emphasis added]

[Update, May 17: Vergano's follow up now has new information on the previously unnamed student who provided the SNA background section for the Wegman report. It's Denise Reeves of Mitre, whose contribution to the report had been acknowledged without specifying her role.

"I was Dr. Wegman's graduate student when I provided him with the overview of social network analysis, at his request. My draft overview was later incorporated by Dr. Wegman and his coauthors into the 2006 report. I was not an author of the report.

"The format of the 2006 report involved a limited amount of citations. The social network material that I provided to Dr. Wegman followed the format of the report."(emphasis hers)

Adding that she has met with a George Mason University misconduct committee, Reeves concluded, "My academic integrity is not being questioned."

Indeed it's telling that the Wegman panel found it acceptable that such a large swathe of background material - five pages - required no citations at all. It should also be noted that the section was considerably reduced when used in the Said et al co-author Walid Sharabati's thesis and the soon-to-be retracted Said et al 2008. And there is really no wiggle room here; if Said et al authors really thought Reeves had produced a totally original exposition of SNA off the top of her head, as far-fetched as that may seem, then she should have received specific acknowledgment in the later work. Instead they put in one desultory citation to Marc Granovetter for one specific statement on the power of "weak ties". ]

And most ironically for a paper purporting to assess the relationship between co-authorship “style” and peer review quality, it is still unclear that any substantive peer review occurred.

Five days after Wegman’s email submitting the paper, chief editor (and longtime friend) Stanley Azen wrote back:

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.

That email was provided to Vergano by Wegman in response to an FOIA last year. Vergano continues with Azen’s comments:

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.

So Azen appears to claim that reviews “must have” been requested and done in the previous week, given the official date of submission, yet he has no electronic records of those reviews and the presumed hard copies have been destroyed.

We’ll give the last word to Virginia Tech’s Skip Garner, who was one of three experts who had earlier confirmed likely plagiarism in the Wegman report and Said et al.

The retraction of an article is a serious and impactful action, for it confirms that a complete analysis by the editors confirmed inappropriate ‘re-use’ of material, and in this case issues with the review process that was in place at the time.

….[I]t is important that the notice of retraction be propagated back to the literature databases and search engines so that future users know not to use the material. Retracting on a web site is only the first step in that process …

And one final note, the finding of ‘plagiarism’ may also be an indicator of other possible questionable ethical issues such as conflict-of-interest, haste vs. scientific rigor and bias, which may need to be investigated.

About these ads

91 responses to “Retraction of Said, Wegman et al 2008, part 2

  1. It’s in Virginia and government money was used and behaviour was bad so maybe Cuccillini can get the emails.

  2. Rattus Norvegicus

    I thought that this was the most devastating pull quote from the Vergano piece:

    “The authors speculate that the entrepreneurial style leads to peer review abuse. No data is provided to support this argument,” Carley says, by email.

    So much for the conclusions not being affected.

  3. Pingback: Why Wegman Matters « The Policy Lass

  4. Ted Kirkpatrick

    Said’s tenure as Associate Editor of CSDA—the only editor listing “Social network theory [sic]” as a specialty—provides an amazing contrast to one of the paper’s key claims. Consider:

    Said Ph.D. awarded early 2005
    Said first listed as CSDA Associate Editor July 2006
    Said first author on CSDA paper submitted July 2007

    Keep the above dates in mind as you read the following sentence from the paper:

    Finally, the mentor style of co-authorship [the "good", Wegman style], while not entirely free of the possibility of bias, does suggest that younger co-authors are generally not editors or associate editors. (p. 2183)

    This sentence is from the paper’s concluding paragraph. It’s not a small point, it relates to the primary claim of the paper: that coauthor networks in the “mentor” style are less prone to gaming than those in the “entrepreneurial” style.

    Self-refuting, all the way down.

  5. Perhaps Azen got an unnamed intern to do the review, using his email account. When he remembers this, there’ll be two unnamed people under that bus. By the time this is over, there’ll be dozens of John and Jane Does under it, but noone else will be in any way to blame.

    All right, I confess. It was all my doing.

    Yours contritely.

    John Doe

  6. Susan Anderson

    Whatever the denialati choose to do, it is great to see the egg that belongs on their faces in the right place for once. They are all too good at pretending but this looks like some of it might stick.

    Here’s hopin’

  7. Pingback: Wegman paper retraction by Journal « Wott's Up With That?

  8. See SSWR, p.151, “Comments on [SAI2008] from an expert.”
    Garry Robins:
    “The literature on network entrepreneurship, including work by Granovetter and Burt, gives plenty of theoretical and empirical reasons to suggest that such groups, linked by one entrepreneur (in this case the central author), may indeed have different opinions or approaches.”

    Basically, he’s saying the fundamental claim of Said(2008) about entrepreneurial networks … is likely wrong. But read the whole thing.

    I could have gotten another quote, but I figured the horse was already dead and needed no more beating.

    Of course, asking Prof. Carley to look at this was a brilliant move on Vergano’s part.

    • Pete Dunkelberg

      I could have gotten another quote, but I figured the horse was already dead and needed no more beating.

      Beware the climate zombies!

  9. Pingback: Controversial paper critiquing climate change science set to be retracted because of plagiarism « Retraction Watch

  10. Wonder what GMU is doing with those three doctoral dissertations that contain plagiarized material? The original allegation to GMU and Purdue was way back in October 2010. I think six months should be sufficient to investigate this matter. I wonder if a follow-up email will produce anything useful.

  11. Deep,

    You and John Mashey got a good cite from John Quiggin at CrookedTimber here:

    http://crookedtimber.org/2011/05/17/adventures-in-social-network-analysis-approaching-the-finale/

    and it was picked up by John Cole at Balloon Juice:

    http://www.balloon-juice.com/2011/05/17/she-blinded-me-with-plagiarism/

  12. Sam: yes, let us hope someone asks.

    Also (everybody):
    If you previously read Vergano’s blog piece, check it again. he updated it with new info about Denise Reeves … she says it’s not her fault.

    I foresee interesting discussions.

    • Gavin's Pussycat

      John, she also says that “she has met with a George Mason University misconduct committee”… suggesting that such a committee exists and is doing work.

      Does that mean that we all misread GMU on this, that instead of dragging their feet, they understand that this is much bigger than a “standard” science misconduct investigation, and are actually intent on getting to the bottom of this, formal deadlines be damned?

    • Don’t forget the process at GMU has two different committees, one at the inquiry stage and one at the investigation stage. As far as I know, Ray Bradley was never advised of the inquiry outcome, as he was promised and as required by ORI rules. But it’s possible that it did move to the investigation; if so, it is very likely that this happened only as a *result* of the negative publicity last fall.

  13. John Mashey writes in his excellent SSWR:

    “Of 91 pages, 35 are mostly plagiarized, but injected with biases, errors or changed meanings that often weaken or invert original results. Some might thus also be called fabrication.”

    As appropriate as it is that folks are beginning to acknowledge Wegman’s plagiarism, let’s not forget the fabrications. Those are even more damning.

    Some people even call them “lies”.

    [DC: There are certainly distortions, especially in the background on tree-rings. It's quite possible, though, that most or even all of these were not deliberate, but rather the inevitable outcome of an extremely biased process and the authors' gross incompetence. That's still unacceptable of course. ]

  14. John:

    he updated it with new info about Denise Reeves … she says it’s not her fault.

    Actually, she said the misconduct committee said it wasn’t her fault.

    I foresee interesting discussions.

    I forsee even *more* interesting discussions … :)

  15. Interest in different places:
    Retraction Watch.

    Chronicle of Higher Education

    In the latter case, a kind reader had given me the name for a contact a while ago, and I tried again. Newshooks are useful.

    People might encourage these folks with a few comments (CHE needs subscription), as very different sets of people read these.

  16. John and Deep,

    Let’s get back to the issue of GMU accreditation. I think this body needs to be informed of lack of progress in the Wegman investigation and the plagiarized doctoral thesis of Said. Apparently, in the Ohio University plagiarism scandal, action was taken only when the accreditation board was informed of the rampant thesis plagiarism.

  17. I do not understand why the editor didn’t send the paper to a reviewer with expertise on Social Networks. If the editor has has no familiarity with a field, he/she sends it to someone who does. It’s not hard to find experts if you google on Social Networks. Such an expert would have noted the weakness of the “research” in the paper.

    Editors seldom review papers; that is the function of the reviewers they send it to. They don’t have time.

    Azen’s claim that the paper had been reviewed is strange. He lost the paperwork, and he can’t remember who would have reviewed it?

    It is NOT a favor to an author to publish a paper that is unreviewed by experts.

    Azen is a very experienced editor. I can’t understand his actions.

    • I do not understand why the editor didn’t send the paper to a reviewer with expertise on Social Networks.

      Because climate scientists need statisticians, but statisticians need no expertise beyond their own.

  18. Excellent example of how there is a major problem with peer review.

    • Not really Grim,

      The bottom line is that good quality, important research is published, and science advances in all fields. Peer review is a strong part of that process. Of course the most fundamental element of peer review is “self peer review”, in other words the essential integrity of scientists for finding stuff out and doing it well.

      A tiny number of individuals lack that basic scientific integrity. They misguidedly consider that the structures of science and scientific publishing can be used for self-promotion or to promote dreary political ideas. They can sometimes get away with publishing dismal rubbish in crappy journals. Wegman and Said are case in point.

      It doesn’t mean that there are necessarily problems with peer review. It’s hard work for everyone if it’s done properly, but it continues to work, plagiarists and nepotists like Wegman, Said, and Azen notwithstanding. Who cares about their crappy journal? We can recognise rubbish and give it the wide berth it deserves…

  19. I do not understand why the editor didn’t send the paper to a reviewer with expertise on Social Networks…

    Azen is a very experienced editor. I can’t understand his actions.

    I’d say it’s highly probable that he doesn’t care for the political implications of the need to do something about global warming, and was more than happy to help his buddy shoot arrows into the collective climate science community’s back.

  20. Steve Metzler

    Excellent example of how there is a major problem with peer review.

    No, it isn’t. It’s an excellent example of what happens when something that is *not* peer reviewed appears in an otherwise reputable journal. Nice try though. Thanks for playing.

    • Steve has said exactly what I was about to say myself.

      I would add that the comment by ‘GrimG’ is itself an excellent example of the abuse of science and scientific process, and of the public perceptions thereof, in a vein similar to how the whole Wegman affair has attempted to turn reality on its head.

      GrimG, remember the quote attributed to Lincoln/Barnum:

      You can fool some of the people all of the time, and all of the people some of the time, but you can not fool all of the people all of the time.

      You’re talking to an educated audience here, and you’re fooling no-one but yourself.

  21. Azen is a very experienced editor. I can’t understand his actions.

    I’d say it’s highly probable that he doesn’t care for the political implications of the need to do something about global warming, and was more than happy to help his buddy shoot arrows into the collective climate science community’s back.

    And now he is certainly paying the price–cf his sad, frantic attempts at damage control. What a cautionary tale for any journal editor tempted to grind an axe in his publication!

  22. 1) There is no evidence of Azens’s beliefs about AGW that I know of.

    2) CSDA actually seems a credible journal – I know some people involved I respect highly. This is not E&E.

    3) It seems more likely that Wegman (note, he was the corresponding author, not Said) sent it to an old friend, who rubberstamped it, perhaps thinking social networks was a hot topic. If you track the dates, Vergano knew about this last Fall, but it seemed fair to let Elsevier, CSDAhave a chance to follow their policies and give Azen a chance to be forthright about this and apologize, in which case I’d have applauded him.

    The result is sad. Azen joins the list of old friends who trusted Wegman and probably wish they hadn’t.

  23. Does anybody know when the retraction will be published? The journal currently has on its website issues up to September 2011 and there is no sign of a retraction. I have written to Prof Azen for guidance but not yet received a reply.

  24. Peer review is imperfect, but is generally effective. What people often forget is that science is a human activity and has always had the typical human biases. Peer review helps to eliminate the biases and expose weak arguments. Peer review gives the opportunity for public criticism, both in the semi-private review and in the comment process after a paper is published. Others observe defects, and eventually eliminate them.

    In other words, sometimes the process is messy but few errors remain after a decade or so.

  25. Bob:
    1) It has been confirmed to several people, but as far as I know, nobody has been told a date. The most recent issue just came out last week.

    2) If you want to know more about what’s going on, send me email:
    JohnMashey (at) yahoo.com will get here. I think there is reason why it has taken this long.

  26. Well done, you guys, for getting some results!

    Meanwhile, we are about to be rescued from ‘post-normal science':

    http://climateconference.heartland.org/

    Those big hearted Heartlanders (sarcasm) also have a wiki:

    http://www.climatewiki.org/wiki/Main_Page

    Hat tip: http://initforthegold.blogspot.com/2011/05/more-yucks-from-denialati.html

  27. Holly Stick,
    thanks for that … I had to laugh when I say who was top of list of scientists attending the conference … Tim Ball!! The man has been mired in controversy about his credentials and is currently being sued by Michael Mann.

  28. Holly Stick:

    Any idea if there’ll be anything new and weird in the coming Heartland ‘conference’? It’s become so repetitive that I stopped keeping track…

    * * *

    While we’re on Wegmangate, check out what (um) used to be Wegman’s home page: http://webcitation.org/5tNd5RApE

    I’m calling this an ‘etymology fail on multiple levels’.

    frank

    • If there’s any interest in discussing Heartland further, head on over to the new Open Thread #10. It doesn’t have to be all Wegman, all the time (although these latest two posts on Said et al retraction are “sticky” and will still show ahead of the open thread).

  29. 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

  30. Well, it’s bizarre, but I offer the following.

    1) I claimed in SSWR the Wegman effort got rolling with two main missions: break the hockey stick (and IPCC), and cast doubt on paleoclimate peer reivew and independence.

    2) Now, Wegman is a senior academic, with an interdisciplinary emphasis and a huge social network.
    How long would it take him to find an SNA expert willing to work on a high-profile report to Congress? Reese had returned from CMU short course a bit before, from Carley. That’s likely this. I found 2 serious SNA experts happy to do a quick review of Said(2008) for me, a stranger, in a few weeks.

    3) But the problem was that if you got an expert, he would need to be sure they would reach the RIGHT outcome, would not quit in the middle. I cannot imagine any real SNA expert signing off on the WR unless they were otherwise motivated. Among other things, their reputation would be on the line.

    4) But it’s a lot easier to do with one’s own students…

  31. William Happer has a new anti-science essay out. (http://www.firstthings.com/article/2011/05/the-truth-about-greenhouse-gases) I didn’t realize he was on the board of MITRE, a defense contractor where several of Wegman’s former students work.

    http://www.mitre.org/about/bot/happer.html

  32. Happer is Chairman of the George Marshall Institute.
    See CCC, p.62-66.

  33. Re: Stanley Azen–in his online CV from USC, among his appointments is Assistant Dean for Research Integrity. One might assume that plagiarism and peer review procedures both fall under the aegis of ‘research integrity’. Curiouser and curiouser.

  34. Well, I’m sure DC will discuss this soon, but for now, see
    Strange tales and emails, just out.
    You may want to save comments for a thread on that.

    • Gavin's Pussycat

      John, why are you publishing on that illicit toxic waste dump known as desmogblog? Do you owe them? How do you like their comments editorial oversight, if any?

      [DC: I don't think that's a fair characterization, although obviously I don't think it's useful to have such an open comment policy. The fact remains that James Hoggan was the first to detail shady anti-science PR in Canada, with DesmogBlog's exposes on the Friends of Science. ]

  35. Andrew Gelman explains why the plagiarism is such a big deal:

    It’s not that the plagiarized work made the paper wrong; it’s that plagiarism is an indication that the authors don’t really know what they’re doing.

    Wegman has implied that copying “boilerplate” isn’t really plagiarism or, if it is, it’s no big deal, not affecting his scientific conclusions. But that’s not really correct. The work in question is not a theorem that’s true or false, it’s a bunch of statements (an “opinion piece,” in the words of CMU prof Kathleen Carley), and in that case the expertise of the authors is an important contribution.

    The motivation for copying without attribution is clear: it allows Said, Wegman et al. to claim expertise without doing the work required to actually be experts.

  36. An Editorial on this in tomorrow’s Nature:

    Copy and paste
    Nature 473, 419–420 (26 May 2011)

    slow university investigation into serious accusations of misconduct benefits no one.

    your role is strongly cited:

    “That doubts about the 2006 report have resulted in concrete action is mainly down to the sterling work of an anonymous climate blogger called Deep Climate.”

  37. The Nature editorial is here:

    http://www.nature.com/nature/journal/v473/n7348/full/473419b.html

    A nit to pick: the “wave of glee” allegation is distasteful.

    Much more important: many thanks to Deep C and John M for their efforts.

    • Rattus Norvegicus

      I have to say I felt a frisson of joy!

    • Gavin's Pussycat

      Pleading guilty too…

    • Gavin's Pussycat

      …and to elaborate, what’s wrong with happiness at seeing crime meet its punishment?
      To understand climatology denialism, it helps to look at it as organized crime. Not a political viewpoint, even less a scientific one. Organized crime. Psychologically if not legally, and Wegman’s doings are so like those of a petty crook, moving from trespass to trespass as if untouchable, until the fall. And while it’s unsatisfactory that the big bosses go free, we should be happy for the small things in life…

    • I’m not sure it is that organized, and I don’t know what “psychological” crime might be – not that I’m excusing Wegman’s indefensible actions. But I’m glad you raise the issue of the “big bosses”. We shoud not forget Barton, Inhofe and all the rest of in the thrall of corporate interests implacably opposed to concrete action, or even rational discussion, regarding AGW.

    • Gavin's Pussycat

      DC, what I mean is that, while society hasn’t found a way (compatible with the rule of law) to criminalize climatology denialism in the way it has narcotics and (in some states, on some level) prostitution, gambling, guns etc., nevertheless the practitioners know that it’s wrong — or would know if they allowed themselves to stand still and reflect. The damage done by denialism to society and to victims is much like that of the above mob-related activities — also something denialism denies.

      Being a denialist requires a sort of moral numbness, of the kind also found in the tobacco industry — and in power politics. Together with an assumption that, we won’t get caught, they cannot touch us, often with a notion of democratic society and its law enforcement being ‘soft’. And then going from transgression to transgression. And when caught (often after a long string of crimes that were never solved or gave a suspended sentence as a ‘first offender’), the error in the criminal mind is not so much the crime, as being caught for it. And excuses, explanations, and more bull. But never an admission or an apology.

      There is much more, but this is how I see in denialism the workings of the criminal mind.

    • Susan Anderson

      Attributions of innocence reflect kindness and tolerance but are not accurate. I think being aware of the white collar near-criminal aspect of all this is important. So many people, like I did, come in willing to assume others are like those we meet in daily life, largely well intentioned if deluded. It took years of careful work for this inexpert but largely honest nonscientist to establish that there wasn’t any substance at all to the counterclaims. And for those (a large majority) without my limited scientific training and experience and liberal bias, accompanied by fascination with worldwide evidence, the work is not going to get done. These guys are not by any definition innocent though its hard to separate them from their dupes are are innocent if dangerously ignorant.

  38. And, in addition to the well-earned effusive thanks due to Deep Climate and John Mashey, a round of applause to Stevieboy McinTyres, for unequivocally showing what a clueless, uncomprehending, un-deflecting dingleberry he truly is</a

    • I dunno, chek.

      There was nothing clueless about Stevie’s manipulation of his infamous hockey-stick shenanigans. It’s one thing to do some fancy maths or stats work on a data (or other) collection. Lots of people need to do that to elicit the meanings or trends within that data.

      It’s something else again to claim that the results are plain, clear and simple and nothing to do with the several processes you’ve applied to get the result you are bragging about.

    • I find it hilarious how he keeps on defending his friend McKitrick and the hopelessly flawed paper with Pat Michaels. His fanboys don’t even dare to point out that that paper clearly was flawed, and deserves no defence.

      Just like Loehle is handled with silk gloves (or Douglass et al, for that matter). The auditor-audience still doesn’t seem to grasp the clear ideological bias in the auditing.

  39. Pingback: Turn 180: Irish Climate Skeptics - Page 46

  40. I agree generally Adelady – but I was referring in this instance to the thousands of words posted over several “articles” and their comments on McinTyre’s site regarding these falsification and plagiarism charges which have added up to nothing but fact-free bloviating of absolutely no use at all, apart from the chief goal of keeping his congregation ill informed, ignorant and on-message.

    The self-appointed Grand Auditor is anything but.

  41. Here’s is another tidbit, related to SIGMU.

    Read carefully.

    p.10 quotes Dan Vergano’s first story, of which I wrote:

    ‘His story included a later comment that has confused some people:
    “Walsch clarified on Sunday that Bradley’s complaint is under a formal investigation by the university, and has moved past a preliminary “inquiry” to a committee effort.
    GMU policies say: preliminary assessment, inquiry committee and investigation committee, A.1. Stough had many times specified inquiry not investigation. Perhaps he and Walsch were not communicating.’

    Then the Nature editorial this week had:
    ‘Daniel Walsch, spokesperson for George Mason University, says that an internal review of the matter began in the autumn. He cannot estimate when that review will be complete, and, until it is, he says, the university regards it as a “personnel matter” and will not comment further. He adds that the review is still in the “inquiry” phase to ascertain whether a full investigation should be held. “Whether it is fast or slow is not as important as it being thorough and fair,” says Walsch.’

    Vergano checks things, and added a May 26 update to his story:
    ‘Update: GMU spokesman Dan Walsch clarified in the May 26, 2011, Nature journal that the year-old investigation is still in its preliminary “inquiry” stage, rather than a full investigation.

    “In terms of my comments this past fall, my understanding of the internal procedure was not as clear then as it is now,” Walsch says, by email.]’

    Stough had claimed (p.31) that the inquiry committee was formed April 2010, but in email (p.32) first meeting didn’t happen until August. At least Walsch’s current comments are consistent with that.

  42. So let me get this straight. GMU is still deciding whether to investigate? Still?

  43. So if I understand things correctly, Wegman’s main points are:

    1. Mentor-style co-authorship results in less corruption when producing papers.

    2. Mentor-style co-authorship results in less corruption of the peer review process.

    3. Climate scientists should have used more statisticians because statistics lies outside their expertise and their results depend on statistics.

    …but his mentor-style co-authored paper is plagued by:

    1. Corruption of the paper via plagiarism.

    2. Corruption of peer review process via no review.

    3. Lack of review by any actual experts in the field of the subject of the paper.

    And the situational irony of the year award goes to…

    • Jeffrey Davis

      Don’t forget the fact that the statistician missed the cherry-picked data.

  44. The footy I was watching on TV got one sided, so I took a look at Said, Wegman & Sharabati 2009 “Author-Coauthor Social Networks and Emerging Scientific Subfields”, in Data Analysis and Classification

    Said et al.: write:
    “A common property of many large networks is that the vertex connectivities follow a scale-free power-law distribution. This feature was found to be a consequence of two generic mechanisms: (1) networks expand continuously by the addition of new vertices (growth), and (2) new vertices attach preferentially to sites that are already well connected (preferential attachment). A model based on these two ingredients reproduces the observed stationary scale-free distributions, which indicates that the development of large networks is governed by robust selforganizing phenomena that go beyond the particulars of the individual systems.”

    Barabasi & Albert (1999) “Emergence of scaling in random networks” Science, 286(5439) write:

    A common property of many large networks is that the vertex connectivities follow a scale-free power-law distribution. This feature was found to be a consequence of two generic mechanisms: (i) networks expand continuously by the addition of new vertices, and (ii) new vertices attach preferentially to sites that are already well connected. A model based on these two ingredients reproduces the observed stationary scale-free distributions, which indicates that the development of large networks is governed by robust self-organizing phenomena that go beyond the particulars of the individual systems.

    Barabasi&Albert is cited on the next page but there is nothing to indicate the above is a lengthy quote. Said et al also write:

    “Preferential attachment means that the more connected a vertex is, the more likely it is to acquire new edges. Intuitively, preferential attachment can be understood if we think in terms of social networks connecting people. Here an edge from actor A to actor B means that actor A knows or is acquainted with actor B. Vertices with many edges represent well-known people with lots of relations. When a new actor enters the community, he or she is more likely to become acquainted with one of those more visible actors rather than with a relative unknown.”

    Wikipedia (uncited) has from 2007 I think :

    Preferential attachment means that the more connected a node is, the more likely it is to receive new links. Nodes with higher degree have stronger ability to grab links added to the network. Intuitively, the preferential attachment can be understood if we think in terms of social networks connecting people. Here a link from A to B means that person A “knows” or “is acquainted with” person B. Heavily linked nodes represent well-known people with lots of relations. When a newcomer enters the community, s/he is more likely to become acquainted with one of those more visible people rather than with a relative unknown.

    • It keeps on giving and giving and giving and giving…

      Yep, they never ‘knowingly’ plagiarised, if you apparently don’t know that “copy-paste, sometimes make small changes” is plagiarism!

  45. Way to go andrewt!
    I looked at at the last year, but only for being another step in the other SNA plagiarism tree, but it didn’t look like it. (It is actually Said, et al(2010),
    via Springer, another one for them.

    Of interest:
    1) Barabasi is quite famous in this turf, noted in SSWR, p.144.

    Said(2009) miss-spells this as Barabsi.

    2) We have a 2nd hit for the same agencies:
    “The work of Dr. Said is supported in part by Grant Number F32AA015876
    from the National Institute on Alcohol Abuse and Alcoholism. The work of
    Dr. Wegman is supported in part by the Army Research Oce under contract W911NF-04-1-0447. Both were also supported in part by the Army Research Laboratory under contract W911NF-07-1-0059.”
    But we get to add:
    “This work was completed while Drs. Said and Wegman
    were Visiting Fellows at the Isaac Newton Institute for Mathematical Sciences.” (that’s at Cambridge…)
    And by then, Sharabati is at Purdue.

    3) Said(2009) has 14 references, all of which are found in Sharabati’s dissertation. See SSWR p.152, [#] is the reference there. The 3 *’d ones are references in Sharabati, but uncited.

    1.[2] Barabsi 9sic), but spelled correctly in dissertation.
    2.[6] Borner
    3. [8]Carley [the CMU expert quoted by Vergano]
    4. [11]Cioffi-Revilla
    5. [12]CIS
    6. [13]FBLP
    7.[18]*Eaton, Kumar, Reingen … which is actually
    Easton, Ward, Kumar, Reingen. Sharabati got that right, although didn’t give reference with a year.
    8. [32]*Klink
    9. [34]Krackhardt and Carley
    10. [35]*Kuriyama
    11. [48]PubMed
    12.[51]Roth
    13.[55]Said et al (2008)
    14.[61]Wasserman and Faust

    4) And unsurprising,
    Said(2010):
    “In this paper, we focus on demonstrating scale-free author-coauthor social
    networks.” … andrewt’s 1st paragraph

    Sharabati, p.143 (of PDF, labeled 128):
    “In this section, I focus on demonstrating scale-free author-coauthor social networks. A common property of many large networks is that the vertex connectivities follow a scale-free power-law distribution. This feature was found to be a consequence of two generic mechanisms:
    (i) networks expand continuously by the addition of new vertices (growth), and
    (ii) new vertices attach preferentially to sites that are already well connected (preferential attachment). A model based on these two ingredients reproduces the observed stationary scale-free distributions, which indicates that the development of large networks is governed by robust self-organizing phenomena that go beyond the particulars of the individual systems.”

    Actually, much of Said(2010) including the graphs seems to come from Sharabati(2008), but it was nice of them to include him as 3rd author.

    Wegman’s resume lists:
    183. “A model of preferential attachments for emerging scientific subfields,” with Walid K. Sharabati and Yasmin H. Said, Proceedings of the Joint Statistical Meetings, 2048-2055, 2007

    200. “Author-coauthor social networks and emerging scientific subfields,” with Walid K.
    Sharabati and Yasmin H. Said, Data Analysis and Classification, (F. Palumbo, C. Lauro, M. Greenacre, eds.), 257-268, 2010.

    SO, looks like another chain: Science – PhD – article

    • John Mashey

      OOPS, not all of this is right, I was looking at a what must have been an older version, with 14 refs instead of 8. I got a copy of the real thing (and at least it spells Barabasi right).

      But andrewt’s 2 chunks are still there.

  46. Steve Metzler

    There is another possibility… that Said originally provided those Wikipedia entries that she is accused of plagiarising. No? Yeah, I suppose you wouldn’t be subtly altering the phraseology then, to make it look like you weren’t palgiarising. My head is spinning…

    • I browsed back through the Wikipedia history over breakfast. The text was added to Wikipedia November 2006 by Francis Heylighen, a well-published professor at the Free University of Brussels.

    • John Mashey

      Andrewt: thanks! That helps.

  47. As usual, it is not yet clear exactly who did what. Who knows what “reading material” was given to Sharabati. Maybe the tale of this one will be as interesting as the last one.

  48. Oops, my earlier post was in error, I';d gotten a copy of what I thought was that paper long ago (I think off Sharabati’s website), but the actual paper @ Springer is slightly different, shorter and 8 references instead of 14, no thanking Newton for support (but newton affiliation still there) Still mentions the 3 Federal agenices.

  49. According to Nature:

    “Tom Benberg, vice-president of the Commission on Colleges of the Southern Association of Colleges and Schools — the agency that accredits George Mason University — says that his agency might investigate if the university repeatedly ignored its own policies on the timing of misconduct inquiries. To get the ball rolling, he says, someone would have to file a well-documented complaint.”

    Would this well-documented complaint have to come from a GMU professor involved in the misconduct inquiry or could anyone file a complaint? It’s not like some big-shot Attorney General is going to materialize, cut into the line, and piously declare that government money was used to perpetrate a fraud.

    Perhaps Dr. Mashey could file a well-documented complaint, but the complaint is really simple: GMU is violating its own policies about the timing of misconduct inquiries.

    Still, there could be legitimate reasons for the delay. It might not just be embarrassment or political pressure. Perhaps the Southern Association could meet with the professors who are working on the inquiry and determine if there is a valid reason for the delay. The GMU spokesman mentioned there was some litigation.

    I think professors are usually trying to do an honest, thorough, fair job when they have to serve on academic misconduct committees.

  50. The complaint to the accrediting agency should be short and to the point and coming from Ray Bradley would be best. The stuff that DC and John have dug up should be appendices, and they should be edited down to show only the most rock solid stuff related to GMUs lack of investigation. The accrediting agency will not be interested in whether there was or was not plagiarism or bad behavior. They will concentrate on whether procedures were followed. The actual information on the plagiarism will be only of secondary importance to them at this point.

    The nature of these things is that the people receiving them hate it, and look for reasons to conclude that they don’t have to do anything. Oh yes, attaching the Nature editorial is also important.

    By this delay GMU has put itself in the position of not being able to simply quash any investigation at the inquiry stage thinking that the only screaming would come from the blogs. It is now Nature, the ASA, USA Today and more.

  51. I’d like to add my warmest congratulations to Deep Climate and John Mashey that their excellent work appears to be, albeit slowly, reaping the rewards it deserves. I’m another one who definitely without guilt felt a tinge of glee that this mud appears to be sticky (high clay content?), and attaching itself to at least some of the deserving recipients.
    The exposures of plagiarism and the discovery of the deliberate errors in McIntyre’s code (shown to be uncritically replicated by Wegman), rank as superb achievements in the defence of science.

  52. Ted Kirkpatrick

    The Rabett is wise. I wonder, though, if there’s an additional angle for getting Southern Association of Colleges and Schools (SACS) ‘s attention: The obvious problems with graduate supervision. These problems are not localized to Wegman. Other faculty signed off on these dissertations, despite obvious flaws in writing, logic, and citation.

    For the plagiarism complaint, SACS would only monitor progress, especially given that ORI will also be reviewing it. But might SACS be more proactive about indications that students are graduating with inadequate training? This cuts close to the agency’s mandate as a guarantor of academic standards. If standards are slipping at a university accredited by SACS, SACS is failing in its core mission.

    If someone has already submitted a complaint explicitly about problems with one or more dissertations, and the process has drawn out, it might be more effective to ask SACS to investigate the progress on that complaint.

  53. I downloaded Sharabati’s thesis and as John Mashey says it also has the text from Wikipedia twice (p30&p128) and the text Barabasi & Albert (p128). I came across another piece of cut-and-paste ( below). I feel an obligation to inform GMU but reluctance given poor supervision maybe the cause.

    Sharabati 2008 (p8&p124):
    MDS represents the patterns of similarity or dissimilarity in the tie profiles among actors (when applied to adjacency or distances) as a map in multi-dimensional space. The map lets us see how close actors are, whether they cluster in multi-dimensional space, and how much variation there is along each dimension. The goal of MDS is to minimize stress – distance between nodes. Stress is a measure of badnessof fit; 0 <= stress 1 <= The range of solutions with more dimensions is sought, so that the analyst can assess the extent to which the distances are uni-dimensional. The meaning of the dimensions can sometimes be assessed by comparing agents that are at the extreme poles of each dimension."

    Hanneman, "Introduction to social network methods" (sited once 15 pages earlier in the thesis):

    MDS represents the patterns of similarity or dissimilarity in the tie profiles among the actors (when applied to adjacency or distances) as a "map" in multi-dimensional space. This map lets us see how "close" actors are, whether they "cluster" in multi-dimensional space, and how much variation there is along each dimension. "Stress" is a measure of badness of fit. In using MDS, it is a good idea to look at a range of solutions with more dimensions, so you can assess the extent to which the distances are uni-dimensional. … The "meaning" of the dimensions can sometimes be assessed by comparing cases that are at the extreme poles of each dimension.

  54. andrewt: argh, yet more.

    I’ve made a pass over Said, Wegman, Sharabati,
    comparing it to Sharabati’s dissertation.

    How often does a dissertation co-supervisor take a dissertation, extract some chunks of text and many graphs, make minor edits in a few places, and stick her and the other co-supervisor’s name in front of the student’s?
    (There is ~zero new substance in the article, and the editing doesn’t even always improve it.)

    I have never heard of a student suing his school for bad supervision, but one wonders.

  55. I feel sympathy for those students who may not have been adequately mentored. But for their supervisor(s), I have no sympathy whatsoever.

  56. The student to feel most sympathy for is Denise Reeves. There are 4 papers, 4 theses and 1 congressional report with plagiarism. Wegman is lead author, 2nd author or supervisor on all. The other authors are all current or former students.

    So who has Climate Audit just blamed: Denise Reeves – which is disgraceful.

    Denise Reeves is not an author of any these publications – she gets a acknowledgement (briefly) only in the report – and some of the publications either predate or post-date her time at GMU . By her account in USA Today she has done nothing wrong & her conduct is not being investigated – and there is nothing to contradict this.

    Hopefully it won’t be given any credence because its buried in a defence of Sharabati based on the bizarre contention you can’t plagiarize when describing others’ work in the introductory chapters – but of course papers get retracted and academics have even been dismissed for just this type of plagiarism. And sadly Sharabati has cut-and-pasted text into later chapters as well.

    • Mosher at CA — “[Reeves] is the plagiarist for the SNA material in the wegman report.”

      Oooooh. I suspect Reeves isn’t the type to take such things lying down given she got through the academic grilling and came out squeaky clean, and has a career ahead of her. I suspect Mosher may have bitten off more than he could even dream of chewing.

    • I am really confused by the claim that Denise Reeves “… is the plagiarist for the SNA material in the wegman report”. She wasn’t listed as an author, so how can she be the plagiarist?

      Furthermore, it appears that most of the only original work in the Wegman Report was actually done by Denise Reeves. The SNA analysis was done by her, even if this work was poorly done. The rest of the WR was mostly assembled from unattributed work of others, or “fake” work that was supposed to replicate the statistical analysis. The statistical analysis was apparently simply a summary of work done by McIntyre. Much of the rest of the report was plagiarized from Bradley and others, and doctored up to suit Wegman.

      It appears to me, that the only original work in the WR was done by Denise Reeves! Do I understand this correctly?

    • No, the co-athorship “analysis”, such as it was, was done by others. So far, Reeves is only credited (after the fact) with what Wegman et al claims to have thought was “original” “boilerplate” SNA background (section 2.3).

      Still unanswered is who wrote the other 32 pages of lightly edited “copy-paste” background material, including the background sections on paleoclimatology proxies and PCA and noise models. Or how the actual named authors can possibly escape responsibility for the SNA background, even though they reused it verbatim without giving Reeves credit in subsequent papers.

  57. And I see Mosher is back to flying his true colors over there, andrewt.

  58. Eli Kintisch (2 June 2011) “Journal Retracts Disputed Network Analysis Paper on Climate” ScienceInsider (AAA).
    Links to deepclimate.org/2009/12/22/wegman-and-rapp-on-tree-rings-a-divergence-problem-part-1/ (under alleged to contain).

  59. Pingback: Episode 20: Show Notes — Irregular Climate

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