Retraction of Said, Wegman et al 2008, part 1

It’s been a long time coming, but there has now been an official finding in at least one of the complaints concerning the dubious scholarship of GMU professors Edward Wegman and Yasmin Said. According to Dan Vergano of USA Today, the journal Computational Statistics and Data Analysis  (CSDA) has officially confirmed that Said, Wegman et al 2008, a follow up to the infamous Wegman et al report to Congress, will finally be retracted following complaints of plagiarism and inadequate peer review.

The CSDA paper, Social Networks of Author–Coauthor Relationships, was a follow up to the 2006 report to congress by Wegman, Said and Rice University professor David Scott. Both the Wegman report and Said et al claimed that the “entrepreunerial” style of co-authorship in paleoclimatology demonstrated lax peer review in the field, while the “mentor” style of an established professor collaborating with former students would be less problematic. All three of Wegman’s 2008 co-authors – Said, Walid Sharabati and John Rigsby – were former or current students

I first discovered apparent plagiarism in the Wegman report in December 2009. I later documented massive cut-and-paste in the social network analysis background sections of both the report and the CSDA paper in April 2010. At the time, I pointed out that both Wegman and Said had acknowledged federal funding from research offices associated with the Department of Defence and the National Institute of Health.

And I also noted that the paper had sailed through from submission to acceptance in a mere six days, suggesting that it had not been properly peer reviewed at all. That astonishing fact and the deeply flawed analysis belied the paper’s central premise; indeed, as John Mashey has noted, this is a prime example of self-refuting paper.

Wegman and Said claim to have done nothing wrong, according to Vergano’s report.

“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.

The original SNA background section in the Wegman report was about five pages and was reduced by about half for the CSDA paper. So even this bizarre explanation admits that the CSDA section was appropriated without credit or attribution. It is also telling that none of the authors were sufficiently well-versed in SNA to produce a background section on the techniques employed.

And I recently showed similar cut-and-paste scholarship in the WIREs Computational Statistics article, Color Theory and Design (see part 1 and part 2). Not only was this article written by Wegman and Said alone, they also are editors of the journal (along with the afore-mentioned David Scott), raising once again the issue of lack of peer review of the pair’s scholarship.

There will now be even more pressure on George Mason University to finally resolve the myriad misconduct complaints launched against Wegman and his team. Not to mention increased calls for the retraction of all the other examples of the pair’s dubious scholarship – including the Wegman et al report to Congress itself.

About these ads

91 responses to “Retraction of Said, Wegman et al 2008, part 1

  1. Finally! But expect the deniosphere to start a major howling fest, wondering why ‘accidental’ plagiarism (seriously, Wegman+Said, baaaad excuse blaming a student) leads to retraction and (supposed) flawed papers are not retracted. Note, I do not wonder. Neither will I wonder that they will only go after Mann, and leave e.g. Lindzen & Choi faaaaaar away. Or M&M05 for that matter…

  2. John Mashey

    Good, let the denialsphere howl.
    In the real world, the paper gets retracted, and that’s just the start.

  3. Ted Kirkpatrick

    Wegman’s explanation doesn’t get them off the hook, even taken at face value. The introduction to the paper doesn’t answer any of the questions a scientific introduction should do. Whether derived from student work or not, the authors should have modified it to state their research question, justify their method of answering it, and so forth. The authors should have made those points, and the reviewers should have required it of them. That no one did says much about how the paper was written and reviewed.

    Wegman’s story also raises big questions of attribution. A student wrote the material comprising an entire section, about 1/7 of the whole paper, and they weren’t listed as authors, or even named in the acknowledgements? Really?

    • I think Wegman tries the “two wrongs is a right” excuse, there: “Oi, we only took it from an unattributed source who actually, we now find, plagiarised it. That’s not plagiarism by us, then!”

  4. Steve Bloom

    Well *now* I finally get it: It’s better to co-author with students since they’re ever so much easier to throw under the bus.

  5. I agree with Ted’s comments on the lack of attribution. It is clearly dishonest to ask a student to write a section of a paper for publication and then not acknowledge them as a co-author. As a university lecturer myself, I regard it as the lowest form of cowardly acts for a lecturer or professor to use their students in this way given the power imbalance in the relationship. But I don’t necessarily think this is what in fact happened here. This explanation could also be simply a way of the authors deflecting blame for their own misconduct onto an anonymous “student”. Either way, it is not acceptable conduct.

    • It is clearly dishonest to ask a student to write a section of a paper for publication and then not acknowledge them as a co-author. … But I don’t necessarily think this is what in fact happened here. This explanation could also be simply a way of the authors deflecting blame for their own misconduct onto an anonymous “student”.

      Even if it were acceptable to use a student (it isn’t) and even if it were acceptable to fail to acknowledge them (it isn’t), surely Wegman & Said had an obligation to thoroughly scrutinise what had been done by this third party, particularly given that the central tenet of their paper was allegations of “lax review”. So either Wegman and Said didn’t bother reading the work (bad), or they did and… (badder).

      Either way this is disgraceful behaviour in and of itself, let alone considering these actions in the context of alleging slack review methods on the part of others.

      Kudos to DC and John Mashey – I hope GMU hangs Wegman & co out to dry.

  6. Congratulations! I wondered if this would ever happen.
    I will read all this carefully.

  7. Lotharsson

    If the student wasn’t listed as an author, doesn’t that leave Wegman et al entirely on the hook, and thus swinging in the breeze?

    • Yah, but who believes this :
      “But, but, it was an anonymous student who did it!! Really, please believe me, it’s true…!!”
      Anyway?

  8. Watts is attacking Deep Climate for “being able to attack a man in another country”. Shocking, sir, shocking! Another country? Have you no shame? ;)

  9. Pingback: The Climate Change Debate Thread - Page 712

  10. Any way you cut it, this is a major embarrassment for the much-vaunted authors of the Wegman Report. Throwing a conveniently-un-named student under the bus, not giving said bus-flattened student proper attribution in the first place, failing due diligence by not checking now-squashed-flat-conveniently-unattributed student’s work…

    I’d get my container of popcorn out to watch the skeptics spin this one, but I suspect there won’t be howls and denials. More like silence or diversion, “nothing to see here, move along”…

    • Lotharsson

      If I were said un-named student, I’d be wondering what my legal options were.

      Just sayin’…

    • Lotharsson, you sure if you were the unnamed student that you want to be named? While you can cry “foul” over what Wegman did, your name is now directly linked to “plagiarism”. Much better to just be able to deny Wegman was referring to you.

    • Lotharsson

      Hmmm, good point, Marco.

      I reckon chances are high that the name of said student eventually makes the rounds, at least on the insider networks at first, and maybe wider…and who knows, there may be still more to the story if their side ever gets told. But their best (and cynical) current move is probably to sit tight and hope their name does not become common knowledge.

  11. Pingback: Collide-a-scape » Blog Archive » Collide-a-scape >> Wegman Paper Retracted, Watts Growls

  12. It’s worse than you think. Wegman just admitted he committed two serious ethical violations. First that he appropriated the work of another (the student to be named later) and second that he did not fulfill the obligations of an author to take part and check all parts of a submission. You could get canned for that.
    ————————————————
    “The co-authors of a paper should be all those persons who have made significant scientific contributions to the work reported and who share responsibility and accountability for the results. Other contributions should be indicated in a footnote or an “Acknowledgments” section. An administrative relationship to the investigation does not of itself qualify a person for co-authorship (but occasionally it may be appropriate to acknowledge major administrative assistance). Deceased persons who meet the criterion for inclusion as co-authors should be so included, with a footnote reporting date of death. No fictitious name should be listed as an author or coauthor. The author who submits a manuscript for publication accepts the responsibility of having included as co-authors all persons appropriate and none inappropriate. The submitting author should have sent each living co-author a draft copy of the manuscript and have obtained the co-author’s assent to co-authorship of it.
    —————————————————–

    http://pubs.acs.org/userimages/ContentEditor/1218054468605/ethics.pdf

  13. Ted Kirkpatrick

    The last sentence of the “Acknowledgments” section of the retracted paper: “The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Alcohol Abuse and Alcoholism or the National Institutes of Health”. [Emphasis added.] Seems to be quite a flexible notion of responsibility.

  14. Dr. Jay Cadbury, phd.

    Whatever at least this Wegman guy didn’t try to make a fake graph [edit ... ]

    [DC: Let's stop right there. But thanks for reminding us of the main contrarian defence - baseless, unsubstantiated and false accusations. Now take it somewhere else. Thanks! ]

    • Steve Metzler

      Whatever at least this Wegman guy didn’t try to make a fake graph…

      How right you are there, Dr. Jay Cadbury. But not for the reasons you imagine. Turns out there was very little original work in the Wegman Report, both in a literary and statistical sense. He just borrowed the cherry-picked fake graphs from Steve McIntyre (made with overcooked persistence), without even vetting his statistical methodology. You really should read this:

      Replication and due diligence, Wegman style

      But you won’t. It will be the eventual undoing of the AGW denial house of cards.

  15. Pingback: The WonkLine: May 16, 2011 «ScrollPost.com

  16. Pingback: Retraction of Said, Wegman et al 2008, part 2 | Deep Climate

  17. notjonathon

    Recent revelations (recent to me, anyway) of a connection between George Mason University and the brothers Koch make me wonder whether the university will ever complete its investigation of Wegman. Given the evidence presented so far, he should have been punished long ago. Perhaps the only thing that might move the university would be an investigation into whether it deserves accreditation. Is there a process which permits you to take the case to the Southern Association of Colleges and Schools, Commission on Colleges? Are there any possible penalties for a school that fails to take corrective action against a faculty member who publicly fails to meet minimum standards of scholarship?

    • Ted Kirkpatrick

      Note how Dan Vergano carefully opened each of his stories with the phrase “federally funded study”. The organizations monitoring use of federal research funds (such as ORI) have much stronger penalties available to them than anything SAC/SCOC could assess. And they’re watching this.

  18. DC,
    A sincere thank you for your persistence and hard work. You put people whose job it is to allegedly identify, investigate and rule on these matters to shame.

    I hope that you receive the well-deserved credit for all your work on this.

    Kudos too to Dan Vergano– a diamond in the rough.

  19. Pingback: Why Wegman Matters « The Policy Lass

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

  21. Congratulations to you and John Mashey for unmasking this monstrous fraud. But where, oh where, are the spineless hacks of the main stream media on what what clearly a putrifying story from the get go? “Who pays the watchers?” seems to be more relevant than “Who watches the watchers?” nowadays. We need more Dan Verganos.

  22. Pingback: Climate Science Watch Update – May 16, 2011 | Climate Science Watch

  23. Pingback: Wegmangate: first blood

  24. John Mashey

    USA Today editorial today is:

    http://www.usatoday.com/news/opinion/editorials/2011-05-16-Report-puts-climate-change-deniers-in-hot-seat_n.htm

    “Coincidentally, USA TODAY’s Dan Vergano reported Monday, a statistics journal retracted a federally funded study that had become a touchstone among climate-change deniers. The retraction followed complaints of plagiarism and use of unreliable sources, such as Wikipedia.

    Taken together, these developments ought to leave the deniers in the same position as the “birthers, …”

  25. DC and JM, kudos for your stellar work here and the critical public service that has resulted from it. The corruptive disinformation campaign on this vastly consequential issue has gone on for long enough, and you both really answered the call here.

    On a less sunny note, while this may not a great surprise to you years after you discovered and reported on all this malfeasance, I wouldn’t hold my breath on George Mason choosing principles of academic or any other kind of integrity over payola on this issue. The sugar daddies on whom they depend to a great extent for sponsorship are not known to concern themselves with public relations where their wallets are concerned, e.g. http://www.tampabay.com/news/business/billionaires-role-in-hiring-decisions-at-florida-state-university-raises/1168680, and on this issue, the stakes couldn’t be larger for their enterprise.

    Make no mistake about it, the Koch brothers are thoroughly dispicable would-be if-they-could-be despots. You’d have a better chance of shaming Mugabe or Quadafi. Add this to the fact that George Mason’s other benefactors almost exclusively consists of fellow travelers and the results are quite predictable.

    • I’m sure that the GMU administration would like to sweep it all under the rug. But perhaps some of the faculty and students are getting other ideas and are tired of seeing their institution’s reputation dragged through the mud because of the actions of two rogue professors.

  26. Ted Kirkpatrick

    Small correction, DC: One rogue professor and one rogue postdoc.

    • I think I was misled by this.

      Officially:
      Said, Yasmin H.
      Research Assistant Professor, Computational & Data Sciences

      According to the bio in the WIREs Comp Stat Guide for Authors, dated December 2010, she is (or was) more than just a “research assistant professor”, but rather co-director of GMU’s Center for Computational Data Sciences.

      The admittedly outdated bio reads (with my annotations in []):

      Dr. Yasmin H. Said is a Visiting Fellow at the Isaac Newton Institute for Mathematical Sciences at the University of Cambridge in England [surely not for years on end] and is a National Research Fellow from the National Institutes of Health.
      … Dr. Said is also the Statistical Methodology Director of the Innovative Medical Institute, LLC, and Co-Director of the Center for Computational Data Sciences in the College of Science at George Mason University [no confirmation of this at GMU website]. She is the editor of Computing Science and Statistics [the proceeedings of Interface], is an associate editor of the journal, Computational Statistics and Data Analysis [no longer], serves on the board of the Washington Statistical Society [social functions], and serves on the American Statistical Association Presidential Task Force on Science Policy [no longer exists AFAIK].

      But, anyway, is “research assistant professor”, still just a postdoc? I guess so, if it is not a tenure track position. It’s what GMU calls “term faculty”. Confusingly, though, the titles of Assistant Professor, Associate Professor and (full) Professor can be either tenure-track or term. But Research Assistant Professor and so on are always “term”.

      http://www.gmu.edu/resources/facstaff/handbook/GMU_FACULTY_HANDBOOK_1-1-2009.pdf

    • And what is the Innovative Medical Institute, LLC? Search me, but it appears to be in Washington DC, and is one of three affiliations of one Samer Ellaham in this paper. It looks like Ellaham was a Washington-area cardiologist, who’s moved on to the Cleveland Clinic in Abu Dhabi.

    • Ted Kirkpatrick

      I stand corrected, Deep. If she has an appointment as “Research assistant professor” at GMU, then she in fact qualifies for the title, “Professor”. The limited-term contract doesn’t by itself indicate a lack of depth. Some very good researchers work this way at very good universities. In Said’s case, we have more substantive causes than her title to question her work …

    • I’m getting the impression that there used to be a clearer distinction between “postdoc” and “tenure track” positions, with not much in between. Now many appointments are of the “term” variety but with the same progression in title and responsibility as “tenure track”, if GMU is anything to go by.

  27. Here is a pretty interesting write-up and picture of Dr. Said.
    It even says she is writing a book about the debate of global warming.

    http://www.docstoc.com/docs/13732317/Yasmin-H

    • It even says she is writing copy/pasting a book about the debate of global warming.

      Fixed that for ya!

    • The book is published in 2007 and listed at Amazon. And guess who the primary author is? Yep, it’s DC’s good old friend Edward Wegman…

      http://books.google.com/books?id=IRIGPQAACAAJ&source=gbs_ViewAPI

      “This book recounts the story and facts – in layman’s terms — behind what many believe is a major statistical flaw in recent paleoclimate conclusions regarding global warming, i.e. that the Earth is becoming hotter due to human emissions of carbon dioxide (CO2). Commissioned by the House Committee on Energy and Commerce and the House Subcommittee on Oversight and Investigation, the authors of the book detail time lines, findings, and interpretations that helped shape the misconception behind the effects of global warming as we know it today. Fraught with the potential for damaging political innuendo and inappropriate social networking overtones, the authors steer clear of passing personal judgments in favor of outlining the accepted controversies surrounding the topic, this for historical and reconstructive purposes. The authors’ report to Congress is included in its entirety as an appendix at the rear of the book.”

      This book has been referenced by Pat ‘what a surprise’ Michaels in:

      http://multi-science.metapress.com/content/x327242236qr41t1/

  28. John Mashey

    cynicus: good sleuthing.
    See SSWR, p.65:
    DC found it last year, tried to order it.

    I checked around, found some sites that said it was in stock, 2-3 days, others said it wasn’t published yet. When I tried to order from former … status changed.

    As far as I can tell, the book doesn’t exist.
    Can you give us the page and usage Michaels has?
    That is new, and possibly intreresting info.

    • It’s in the reference list of this paper by Michaels.

      It’s also referenced in the EPA submission by the George Marshall Institute, but the quote is from the Wegman report.

      “Especially when massive amounts of public monies and human lives are at stake, academic work should have a more intense level of scrutiny and review. It is especially the case that authors of policy-related documents like the IPCC report, Climate Change 2001: The Scientific Basis, should not be the same people as those that constructed the academic papers.”9

      9 Wegman, Edward et. al., “Ad Hoc Committee Report on the ‘Hockey Stick’ Global Climate Reconstruction,” Prepared for the House Energy and Commerce Committee, July 19, 2006: 51; see also Wegman, Edward, Controversy in Global Warming: A Case Study (Wiley-Interscience, 2007).

    • Rattus Norvegicus

      DC, the irony here burns, doesn’t it?

    • “good sleuthing.”
      Well, you guys found it much earlier apparently, so who’s the real sleuth here?
      Your (plural) work is much appreciated, I merely hoped to be able to contribute a teeny tiny bit.

  29. @ DC
    “And what is the Innovative Medical Institute”

    Well a quick look at the website and a follow-up google on Ingles-Hospitial suggest something of a Woo , oops I mean complimentary medical institution and probably an off-shoot of a Mexican cancer clinic.
    Any clinic that starts with Celation and Ozone treatments seems a bit dubious. I’d wonder at Said wanting to use such an affiliation.

    • Rattus Norvegicus

      NIH does have it’s own little woo division, The Center for Complementary and Alternative Medicine. As far as I know they haven’t found much woo that woorks.

  30. Mark Shapiro

    Cynicus and Deep,

    I just looked up the Michaels paper you referenced. The title is:
    “Evidence for “publication Bias” Concerning Global Warming in Science and Nature”

    and the first line of the abstract is:
    “The climate research community believes that published findings on global warming will have an equal probability of raising or lowering forecasts of climate change and its impact.”

    Anyone else see a problem with this claim?

    How about:

    1) Where is the evidence for this “belief”.

    2) Michael’s claim assumes that the case for AGW can’t get stronger as more data is gathered, as temperatures increase, as ice melts, as sea level rises, etc.

    Is this a shockingly false premise?

  31. The Wegman book is starting to look like vaporware though it just may be my lack of search skills. If you follow the links at the site cynicus provides none have the book in stock and one or two imply that it is not published yet. One site said it would be out by May 31 2011!

    There seems to be no trace of it on the Wiley site (again could be my poor search skills) and quick searches at the Library of Congress, George Mason University library, Fairfax County Public Library, New Your Public Library, the University of East Anglia (know thy enemies), other universities internationally and various public libraries and a couple of major research libraries (CSIRO in Australia, CISTI in Canada) show no trace of it.

    • Vaporware? Possibly. Very limited circulation might also be possible. How could Michaels cite the book if it didn’t exist? Oh, never mind…

  32. Pingback: Climate Change Denialists vs. SCIENCE — Science Wins! : Delaware Liberal

  33. One of her books listed on Amazon was translated into Arabic and German and published by what has been described as an academic author mill: VDM Verlag Dr. Müller (January 23, 2009) .

    http://www.vdm-publishing.com/

    This blog claims the company is an academic author mill and specializes in “cobbling free Wikipedia entries into expensive books.”

    http://accrispin.blogspot.com/2009/09/victoria-strauss-vdm-verlag-dr-mueller.html

    I gave you some clues…

  34. Amazon spells it Muller (see above) but VDM spells it Mueller.

    Here is the Amazon link.

    Sometimes there are self-publishing companies that charge a lot for books, but you wonder a bit if people are really sending in the money for a book.

    • That appears to be possibly a repackaging of her PhD thesis. If so, the price is quite steep for something that used to be available electronically for free.

    • That might be Said’s PhD thesis. If so, it’s a little steep for something that was available for free electronically until quite recently.

  35. The blog claims (go to their site for links):

    “VDM also does business as VDM Publishing House, LAP Lambert Academic Publishing, Südwestdeutsche Verlag für Hochschulschriften, Verlag Classic Edition (VCE), and Alphascript Publishing–which appears to specialize in cobbling free Wikipedia entries into expensive books, (VDM’s defense of this policy can be seen here). ”

    http://accrispin.blogspot.com/2009/09/victoria-strauss-vdm-verlag-dr-mueller.html

    Maybe the book you can’t find by Said was published by the VDM.
    This company charges a lot for the books it sells, but the authors are paid very little.

    • No the book I was looking for was by Wegman and supposedly published by Wiley. I don’t think it ever came out. But maybe Pat Michaels had an advance electronic copy.

  36. John Mashey

    Oh, I think it was vapor, like Wegman’s promise to put up code/data on website.

    In any case, Wiley dodged a copyright bullet by not publishing this, given that a big chunk would have been text of “pro bono” Wegman Report.

  37. Here is the Wikipedia entry on “author mill.” They mention that the blogger Victoria Strauss coined the term:

    “Author mills don’t require authors to make any financial expenditures at all, hidden or otherwise. However, they do rely on their authors as their major source of income (through books purchased by the author for re-sale, or sold to “pocket” markets the author him/herself is responsible for identifying), and so can be defined as vanity publishers, despite the lack of upfront or other charges. Also, author mills tend to share a business model with vanity publishers: no editorial screening of submissions, no meaningful pre-publication editing, no meaningful post-publication marketing or distribution. ”

    Typically an author mill does the cheapest possible job of production; it sets high cover prices and prints its books “on demand.” The books are listed with on-line booksellers such as amazon.com and bn.com, and on the publisher’s website. Any marketing, promotion, or physical bookstore placement is up to the authors themselves. While authors are not “required” to buy any of their own books, authors who wish to find readers discover that they need to buy their own books for resale.”

    …Victoria Strauss has used the examples of PublishAmerica and VDM Verlag Dr. Mueller to illustrate the concept of author mill.[1][3] More precisely, she has characterized VDM as “an academic author mill”.[3]”

    http://en.wikipedia.org/wiki/Author_mill

  38. Dr. Wegman’s disappeared book is titled “Controversy in Global Warming: A Case Study in Statistics” (John Wiley & Sons, 2007 ).

    http://books.google.com/books?id=IRIGPQAACAAJ&source=gbs_ViewAPI

    The book Dr. Said is “currently writing” has a very similar title: “Controversies in Global Warming–The Heated Debate: Case Studies in Statistics.”

    What I wonder is if this is pretty much the same book.

    • Yes, that’s a good catch. The plural title suggests that the concept of the book may have been expanded to cover more than just the MBH hockey stick papers.

  39. There are two links that mention “Controversies in Global Warming–The Heated Debate: Case Studies in Statistics,” Dr. Said’s upcoming book.

    Like I say, the title is a lot like the Wegman book published by Wiley in 2007 you can’t seem to locate.

    Here is one:

    http://www.docstoc.com/docs/13732317/Yasmin-H

    A University in the United Arab Emirates called the University of Sharjah posted the same biography of Dr. Said in connection with a United Arab Emirates 2008 Math Day, although the photo seems to be dated 2009.

    http://www.sharjah.ac.ae/English/Conferences/uaemathday08/Documents/Yasmin_H_Said_Bio_Photo_03-18-2009.pdf

    Perhaps you could find out if Dr. Said’s book was/will be published in Arabic.

    The University of Sharjah site actually has a Bachelor of Science program in sustainable and renewable energy. They have windmills on their site.

    It looks like a really beautiful campus.

    http://www.sharjah.ac.ae/English/Pages/default.aspx

    “While rooted in Islamic and Arabic traditions, the University of Sharjah promotes an international outlook and respect for other individuals and cultures. UOS takes pride in being a major international center for learning, culture and education where east meets west and north meets south. We look at learning from international perspective, and our educational programs are linked through agreements with universities in Australia, Canada, the UK and USA.”

    http://www.sharjah.ac.ae/English/About_UOS/UOS/Pages/GlobalVision.aspx

  40. deconvoluter

    Color Theory and Design by Wegman & Said,2011 (Ref.1) vol.3,March/April; (Thanks to DeepClimate for starting the ball rolling on this paper).

    Comments by deconvoluter 21st May 2011.

    Introduction. This mini-review is based on a grey scale copy of the first part of the paper which omits the colours. I have only read as far as the end of page 113 and have postponed reading DC’s part2 .

    Although there is a practical bias, the title which refers to theory is a bit over- ambitious. It turns out that most of colour theory has either been omitted or presented in a superficial and inaccurate manner. The first section called the human visual system is a disappointment because it fails to warn the reader that it is restricted to the first of several stages involved which actually culminate in the colour cells in the cortex.

    Some explanations are quite inadequate e.g. that of areal perspective which omits reference to scattering by air and aerosols, of dichromatism which omits the opponent mechanism, the so called value of object colours which involves a confusion about dimensionality and the poorly described
    purpose of the CIE scheme. The apparently fixed nature of the three primary colours may just be a sloppy writing but it does remind me of a similar ‘banning of rotations’ in the McCintyre/Wegman incorrect criticism of the non-centred PC analysis in the first hockey stick paper by MBH. (Ref.2)

    The paper looks rather old fashioned. It would have been better if the authors had consulted some experts in colour vision.

    Details.
    It might have been safer to plagiarise this :

    http://en.wikipedia.org/wiki/Aerial_perspective

    than to include this from p.110 Col.2, p. 113 , L17

    .. distant objects often appear bluish and somewhat indistinct as, for example , viewing mountains or forests from a distance. This is a consequence the greater refractive index for shorter wavelength of light. This bluish hazing effect is known as aerial perspective and is an experience common to all cultures. This is why the sky appears blue. Similarly , this same refractive index effect is why the moon appears reddish when low in the sky.

    Hiding the dimensionality.
    Four variables are used in this paper, but only three at a time, and not always the same three, so the reader must surely become confused.

    Page 108, col.1,Line 1

    It persisted for many decades, eventually being superseded by systems such as the CIE.

    If it has been superseded then it must have been by a comparable system for measuring the colours of objects such as the natural colour system (NCS). More later.

    If the Munsell system established the three dimensional nature of color perception , the CIE models attempt to establish numerical values.

    .

    But these two systems are not comparable. Amazingly this paper omits to discuss black which is in the Munsell and NCS spaces but absent from CIE. The same applies to brown and grey. Newton
    first discussed these colours. He found that brown looked orange when the surround was removed by looking through a narrow tube. The Munsell scheme refers to the colours of objects or surface colours whereas the CIE refers to disembodied colours or sources of light.

    The gap can be bridged by shining three different coloured lights of variable intensity on to a circular patch of white screen in a darkened room. That is a rough way of reproducing all the colours included in the CIE scheme, but nothing grey or blackish. To simulate surface colours would require an additional white annulus surrounding the disk illuminated by a fourth projector of variable luminance. If the annulus is bright enough it can darken the inner disk and even make it go black.

    Notice that black is created by a bright light.. In general the brightness of the annulus has to be specified in order to determine the colour of the inner disk.

    The discussion of CIE continues by incorrectly defining the purpose of it :

    To measure and predict the appearance of a particular colour..

    [My italics].
    CIE is concerned with colour matching, with the reproduction of colour and with the avoidance of metamerism (see text) but has nothing to say about the appearance of colour which is a large and well researched area not covered in this paper. Coloured versions of the CIE charts can be misleading because colour of the real world depends on the whole scene and is not locally determined.

    Now more on dimensionality: consider the summary on p.110 col.2.

    Value or Lightness or Brightness or Luminosity

    The sense here is that these four quantities are very similar. Ignoring the last rather old term, the other quantities are not all comparable. For example brightness is a sensation correlated to the physical quantity luminance and has no upper limit whereas lightness is a sensation correlated to the physical quantity reflectance which is bounded. The article may confuse them but our perceptions do not. The reason for this is that value and lightness are relative terms involving ratios. This muddle has arisen because the fourth variable (illuminance) has been hidden from the discussion.

    you could also be excused if you find this quote tricky:

    Intensity or Purity : how bright or dull a color appears, also called saturation and/or chromaticity.

    Chromaticity and purity are two and one dimensional quantities respectively. Are we supposed to first replace chromaticity by a scalar magnitude?

    But what about the saturation?
    To discuss it properly would raises another issue which this paper avoids i.e. the distinction between stimulus and response (or sensation). Saturation and hue( usually), for example refer to the latter whereas chromatacity, CIE etc. refer to the former.

    Colour blindness mentions ‘yellow’
    p.113 col.1, Line 14,

    a protanope will have no red receptors

    This is correct and so is this from p.112, col.2

    ….red-green colour blind and generally see only blues and yellows

    This refers to people who lack one of the two most common pigments. The authors have not thought about the trap which they have laid for the reader. Why should removal of a red receptor, leaving only
    green and blue, cause the subject to see just yellow or blue?
    Recall that yellow = red + green
    in the CIE scheme. This paper’s account of colour perception is presented inconsistently by omitting all discussion of what happens beyond the cone cells.

    On p.105 under Color Theory.
    Banning ‘legal’ rotations.

    all color perception in humans can be reconstructed using the three so-called primary colors, red,green and blue

    This is partially correct,except that the use of the definite article suggests that there might be a particular choice of precisely defined primaries.

    Schrodinger ,1920, (ref.3)expresses it unambiguously as follows:

    Select three realisable colours as calibration colours, among which there does not exist any linear relation ,i.e. colours no one of which can be mixed from the two other colours [he was referring to coloured lights].

    Now it is clear that these calibration colours are not unique *. Changing from one set to another involves an affine transformation (section 8 of Schrodinger,1920) or roughly speaking a rotation and deformation. In contrast the above wording from this paper encourages the reader to ignore the possibility of this change of coordinate system. The original CIE system does use arbitrarily constructed red,green and blue standard primaries but the version displayed in Figure 9 on page 109 uses quite different primaries.

    The failure to explain or even mention this, is analogous to the attempt by Stephen McIntyre to do the same in PCA space when he incorrectly condemned MBH (ref. 4) for rotating their basis vectors

    Other comments.

    p.108, col.1, Line 8 up;

    The human eye to respond to three different wavelengths is called tristimulus .

    Wrong! (see Figure 8). Is this a consequence of the unfaithful plagiarism discovered by DC?

    Page 108, Col.1,Line 11.

    Scientific values were established earlier this century

    i.e. the twentieth. When did this text first appear?

    Conclusion and summary. A good source, such as the simplified one in Feynman’s lectures in Physics, conveys a sense of security.The reader knows that (s)he is in good hands. Readers of this paper
    should be warned to check every definition and every explanation.

    This is a not very successful attempt to simplify and narrow down a large subject. There are other simplified accounts which are easier as well as more reliable.

    The narrowing is rather severe. There is no mention of opponent colours, or the related NCS colour scheme based on it, or of the coloured shadows experiment, Land’s two colour projections, Zeki’s discovery of two distinct types of cells in the brain devoted to wavelength and colour respectively.

    References (incomplete)

    1. Wegman & Said,2011, vol.3,March/April; p.110 Col.2, p. 113 , L17 (Thanks to DC for referring to this paper)
    2. Tamino at
    (a) http://web.archive.org/web/20080415003018/tamino.wordpress.com/2008/03/06/pca-part-4-non-centered-ho
    ckey-sticks/

    (b)http://www.realclimate.org/index.php/archives/2010/07/the-montford-delusion/

    3. Schrodinger ,1920. Ann. der Phys.63,p.397.

    4. Wikipedia
    ————————————-
    ————————————-
    * In contrast to the colour matching theory considered in this paper, the colour opponent theory works with principle hues which are fixed for each person and which are the basis of the NCS scheme. These perceptual hues are called red,green,blue and yellow except that this red would be classified as purple in the cie stimulus based classification. These are shown at the four poles of the compass here: http://en.wikipedia.org/wiki/Natural_Color_System

  41. deconvoluter

    Corrections:
    the missing line at the end is here :
    These are shown at the four poles of the compass here: http://en.wikipedia.org/wiki/Natural_Color_System

    Also: Part of the reason for the mess is that the italic tags within the blockquotes seem to be italicising the complete contents of the blockquotes.

  42. John Mashey

    Good statistician gets interested:
    Statistics in the news – but not in a good way.

    • Lotharsson

      Mosher on that thread:

      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.

      Mosher’s argument depends on Azen actually having reviewed the paper; it also fails to note that Wegman suggested leading researchers work with their own students instead of other leading researchers.

    • Steven “Piltdown Mann” Mosher on that thread also – quite subtly – tries out the current denialist meme that the plagiarism is unimportant to the Wegman Report because the follow-on paper only takes up the SNA work, not the original hockey stick paper.

      He says, sounding oh-so-innocent:

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

      The statistician who owns that blog says essentially the same thing as the statistician interviewed for the US Today article:

      The flaw is that there’s no there, there to back up their conclusions regarding peer review and social networks.

    • Mark Shapiro

      John – a quick FYI –

      The link to your Wegman analysis that you put on the StatisticsForum is broken — missing the “f” in pdf.

    • However, I notice that you’ve provided the correct URL over there, so John doesn’t have to.

  43. deconvoluter

    Pages 114 to end of Wegman and Said Colour paper.

    Most of the ‘theory’ is now out of the way. But it is still a bit hard to read. When in trouble go to DC’s Part 2, most of the time… I have several, but shall limit myself to one, comment.

    The trouble with referring to the past is that they only got some things right. The text should have helped us, but has it?

    Page 115 last para.

    .. the RYB color wheel. The RYB color wheel was orignally described by Isaac Newton in his 1704 treatise on Opticks. The 18th. century understanding of color vision was predicated on the idea that red,yellow and blue were the primary colors.

    Is the term color vision appropriate when ideas about colour were mainly attributed to the light before it enters the eye i.e. to physics rather than physiology?
    The text is ambiguous, but the suggestion that Newton believed in just three primary colours, as defined earlier in this paper, should have been ruled out explicitly but hasn’t been:

    Whether it may be compounded of a mixture of three taken at equal distances on the circumference ,I do not know ,but of four or five I do not much question but it may [Refers to white]

    Newton’s Opticks. Prop. VI Problem II.

    Also the original version of Newton’s colour circle looks different and has a different function from the RYB colour wheel in Fig.16 (same order, very different proportions).

    The term primary colours has been introduced earlier in the paper. Without warning, the authors have switched to an undefined 18th. century usage. We should have been told, and are not, whether it refers to lights or paints. Historically this was sorted out in the 19th century. But what about the account of RYB in this paper? We are still unclear whether this is based on any kind of mixing because we have not been told. It affects the next topic.

    Complementary colours. Do these add to produce white , as on page 107 or black as in a subtractive RYB scheme or do they merely refer to colours placed in diametrically opposite positions in a colour scheme?

  44. John Mashey

    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, the first 3 lines of the table show papers at 5 or less days, median is 254, and no othre took less than 61.

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

    BTW, I’ve fixed some typos and made a few other minor improvements to Strange tales and Emails.

    I got a succinct comment from the unnamed SNA expert I’d asked this summer, when they heard about the retraction:

    “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!”

    This is sad, but has much truth. I wish I’d written it myself.

  45. steven mosher

    Dehog,

    That’s an starightforward question. For sake of clarity I think its important to get a clear picture of Said 2008. When we finish that we can move onto the balance of the wegman report.

    The provenance of Said 2008 now seems clearer. In John’s side by side analysis he did not seek out a reference for the first paragraph of section 1. We now know that Sharabati wrote section 1. We know he took it from his dissertation. We know he copied portions of Wegman for his dissertation. We know that Reeves wrote those sections. So the sources question seems pretty well settled. Reeves->Sharabati->Sharabati. With said and Wegman and the Journal caught responsible. The next logical question in my mind is what exactly did this paper claim? what did it claim to prove. Since you guys have read it more intently than I have, I thought it would be great to get your views on the major claims of the paper.

    Not really that complicated. I’m willing to accept your interpretation as a starting point.

    When That’s all done, then we can turn to the nice work that DC has done on the stats part. But I like to tackle these issues serially. Makes for a cleaner exposition.

    At the bottom however I dont think Wegman or Said could say or prove anything that would shake my belief in AGW. but its of academic interest

    • Not so fast. There are a lot of problems with your “analysis” of the plagiarism. Your CA piece only mentions Wikipedia and omits the more important sources i.e. the two text books including the seminal Wasserman and Faust.

      Actually I surmised a long time ago that Sharabati had done the reduction of the Wegman report section on SNA that appears virtually identical in Said et al and the dissertation (although I wasn’t sure of the order). As lead authors and dissertation advisors, Wegman and Said seem to have a knack for ignoring the obvious – that in all these cases the work could not possibly be original exposition. How can one have five pages of background material without one citation? And your contention that Sharabati did not claim the first 30 pages as “original” and therefore didn’t need to provide any attribution is ludicrous.

      The real problem here is that this “citation-free” style of appropriation runs throughout the Wegman group’s recent work. After all the two other background sections of the Wegman report contain unattributed (and distorted) material on proxies (derived from Bradley) and on PCA/noise models (multiple sources) – not to mention the copy-and-paste “summaries” of “important papers”. And we see this same pattern in the Wegman and Said 2011 on Color Theory and Design. Wegman or Said must have been responsible for all of those – or are there other students to be thrown under the bus?

      The main “conjecture” regarding co-authorship in the Wegman report, supplementary testimony to Rep. Stupak, and the later Said et al, is that the so-called entrepreunerial style of co-authorship in paleoclimatology would lead to lax peer review, while Wegman’s own “mentor” style would be less problematic. There is no actual evidence for this of course (quite the contrary in fact).

      Rather than the red herring issues of co-authorship styles, or so-called “pal review”, it is“pal editorship” that should be examined, as seen in the example of the Azen “rubber stamp” of Said et al. At WIREs Comp Stat, Wegman and Said take this to new heights by editing their own work and bypassing effective peer review altogether.

    • Not so fast, DC, WIREs Comp Stat has a third Editor who may have been ‘responsible’ for the article…(that’ll be David Scott).

    • The silence from David Scott is deafening. My understanding is that Scott had very little to do with the Wegman Report, beyond the appendix on PCA. And the WIREs Comp Stats overview articles by his co-editors Wegman and Said are truly the “worst of WIREs”. At some point Scott will have to dissociate himself from them, and start cleaning up their mess.

    • Put an end to attempted nick insults like “Dehog”, please.

      [DC: A fair request. ]

    • Mosher

      “We know that Reeves wrote those sections. ”

      Yet, by not making her an author, Wegman is claiming he wrote them.

  46. And note that the “Piltdown Mann” reference was one that Mosher once wore with pride (as it was of his own making) before he decided to reinvent himself as a staunch believer in AGW (who only claims that mainstream climate scientists are dishonest and wegman was merely sloppy and therefore worthy of exoneration …)

  47. John Mashey

    Let’s get timeline and facts clear amidst the obfuscation, referencing STaE and SSWR.

    1) The authors of the WR were Wegman, Scott, and Said, although Scott wrote only the 3-page Appendix A and generally didn’t have much to do with this whole thing.
    That leaves Wegman and Said as the authors of a piece in which 35 of 91 pages have well-documented plagiarism.

    2) The WR has the vague acknowledgement:
    “We would also like to acknowledge the contributions of John T. Rigsby, III, Naval Surface Warfare Center, and Denise M. Reeves, MITRE Corporation.”

    Although Rigsby apparently generated most of the graphs, and Reeves wrote almost pages, *they* weren’t authors, but Scott, with 3 pages, was.
    (See SSWR, p.86, on ASA Ethical Guidelines, especially C.)

    Sharabati was not mentioned.

    3) So, sometime in late 2005 or early 2006, Reeves was asked to write a few pages. See STaE, p.7, notes (4) and (6).

    4) Wegman not only throws Reeves under the bus, but drags in Sharabati as though he had something to do with the WR. Wegman+Said included Reeves’ material in the WR and they certainly knew they got it from her, and of course, as a tutorial for Congress, it is junk, but looks impressive, with dyads and triads, and such.

    5) Said, Wegman, Sharabati, Rigsby (2008) was actually received by CSDA July 2007, and it was October 2008 before Sharabati’s dissertation was accepted (2+ years after the WR in July 2006, and longer after the text was written.) See the flowgraph on SSWR p.118, and especially track the “statuses” to “statues” to “states” silliness. Of course, I would update that to show Antecedents -(Reeves) -internal version that went directly to the other 4 uses (WR (2006), Said(in 2007 really), Sharabati (2008), Rezazad (2009)). Wegman hasn’t said how the text got to Rezazad

    5) But somehow in all this, it isn’t really Wegman & Said’s responsibility, but confused grad students… but the authors, in order are:
    Wegman, Scott, Said: WR(2006)
    Said, Wegman, Sharabati, Rigsby (~2007)

  48. A senior German medical researcher, Peter Zabel, has just had to resign due to plagiarism in a 2009 survey article he co-authored . A survey article of course, describes others’ work and contains no original work.

    Even when you are not describing your original work, it is not acceptable to take others words and depict them as your own, it is plagiarism.

    The unattributed use of others words in Wegman&Said’s 2011 survey article is also plagiarism whatever Steven Mosher says.

    • Note that Mosher has consistently applied the worst possible interpretations to the stolen climategate e-mails and now applies a personal, narrow interpretation of plagiarism that bears little resemblance to that in common use in the research community …

  49. If you want a good example of ‘pal editorship’ causing problems, go back to the Courtillot/IPGP scandal at EPSL in 2008. There the evidence was that three editors from the IPGP (a French national research centre on geophysics) over a number of years were responsible for ‘editing’ an unprecedented number of contributions from their own lab, and even from their own ex-director (Claude Allegre). See here for instance:

    http://news.sciencemag.org/scienceinsider/2008/12/inside-help-for.html

    http://www.nature.com/news/2009/090107/full/457140a.html

    (also the Courtillot page in french Wikipedia).

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

  51. Pingback: Resignations, retractions and the process of science [RealClimate]

  52. Pingback: GMU contradictory decisions on Wegman: Plagiarism in CSDA, but “no scientific misconduct” in 2006 congressional report | Deep Climate

  53. Pingback: James Gentle and Karen Kafadar take over at WIREs Computational Statistics | Deep Climate