Wegman Report

The so-called Wegman Report, written by Edward Wegman, David W. Scott and Yasmin Said, purported to critically examine Michael Mann's "Hockey Stick" "past temperature" reconstruction and its critiques by Steve McIntyre and Ross McKittrick. "Commissioned by Rep. Joe Barton (R-TX) and Rep. Ed Whitfield (R-KY)...[it had been] promoted as “independent, impartial, expert” work by a team of “eminent statisticians.” It was none of those.   As detailed in John Mashey on Strange Scholarship in the Wegman Report, the Wegman report was “a facade for a PR campaign well-honed by Washington, DC ‘think tanks’ and allies, underway for years.”


 * "This supposedly “independent” panel began with a sounding out by a rabid Republican partisan and convinced climate “skeptic”. And Wegman agreed to a process that not only excluded climate scientists, but also involved [Republican Congressional staffer] Peter Spencer as a key conduit and gatekeeper providing climate science documentation and commentary. And all this was done by a House committee that had refused to even acknowledge the offer of a proper scientific review from the National Academy of Sciences."

The Wegman report's statistical criticisms of the Michael Mann's hockey stick finding don't invalidate the hockey stick finding, which has been corroborated by further research and better statistical analysis.

Atypical protocol for such an investigation
Kate Sheppard writes that
 * "[Congressman] Barton's enlisting of Wegman to conduct the report is...outside of typical protocol for committees looking for an independent read on scientific disputes. Congress will generally request a report from the National Research Council, a branch of the National Academies, when there is disagreement on an issue like multiple climate studies."

National Academy of Sciences offer rebuffed

 * "Ralph Cicerone, president of the National Academy of Sciences, wrote to Barton in July 2005 offering the Academies' services. But rather than asking the NRC to review the science, Barton asked Wegman to write the report."

National Research Council: "hockey stick" supported
Sherwood Boehlert (R-N.Y.) objected to Barton's action and asked the National Research Council to study the issue; the Council's report "largely supported the work of Mann and Bradley, noting that while there may have been some justifiable questions about statistical methods and that some uncertainties "have been underestimated," there was nevertheless "an array of evidence" supporting the main conclusion of their work."

WR's evaluation of Mann's "hockey stick" statistical analysis & skeptics' critiques
The "key analysis section of the Wegman Report (WR) on the Mann et al “hockey stick” temperature reconstruction...uncritically rehashed Steve McIntyre and Ross McKitrick’s purported demonstration of the extreme biasing effect of Mann et al’s “short-centered” principal component analysis....in data preprocessing of tree-ring proxy networks".

WR Shortcoming: Report cherrypicked "red noise->hockeystick" simulations
Wegman report accepted McIntyre & McKitrick's assertion that the PCA method Mann had used will create "hockeysticks" just from (pseudo?)"red noise" data, and it reprinted M&M's figure (as Wegman's Figure 4.1) showing a "sample" of simulations generating such hockeysticks - a figure that closely resembles Mann's temperature reconstruction. However, blogger Deep Climate reported that an examination of McIntyre & McKitrick's R script reveals that this similar-shaped "sample" was cherrypicked, drawn from the top hockey-stick-like 1% of the simulation runs.

WR Shortcoming: Report neglected to mention that reanalysis still yields hockey stick
A reanalysis of Mann's tree ring data by Wahl and Ammann (dated 2007, but available online in early 2006) yielded only a minimally different hockey stick ; but Wegman et al neglected to address the question of whether a reanalysis would change the result.

WR Shortcoming: McIntyre & McKitrick omitted renormalization step - and used "inappropriate noise model"
Deep Climate also reported:
 * "the biasing effect [of McIntyre & McKitrick's analysis of "noise" data] was exaggerated by focusing only on PC1 in the data reduction step, instead of ...[Mann's] algorithm end-to-end.

And even within this narrow focus, M&M’s choices greatly exaggerate the [spuriously generated] “hockey stick” effect, i.e. by omitting the MBH renormalization step and using an inappropriate noise model."

SNA 'Outgrowth' journal article retracted
A paper by Said, Wegman and coauthors had grown from the Social Network Analysis section of the Wegman Report, and raised questions about the propriety of climate scientists' co-authorship patterns. Published in the journal Computational Statistics and Data Analysis, its peer review took just 6 days ("submitted July 8, 2007 and accepted July 14...a very fast review...Most take months" .)

In May 2011, the article was retracted: the journal "decided to pull a little-cited 2008 paper on the social networks of author–co-author relationships after it emerged that sections were plagiarized from textbooks and Wikipedia. ... Two of the paper's authors, Yasmin Said and Edward Wegman, both of George Mason University in Fairfax, Virginia, are also authors of an infamous 2006 report to Congress..."

"An opinion piece": retracted article "panned by expert" (USA Today)
In 2011 Dan Vergano of USA Today asked Dr. Kathleen Carley, a social network analysis expert who'd taught social network analysis to the individual who provided material for this section of the Report, to evaluate the now-retracted paper on its substance. In the article Retracted climate critics' study panned by expert Vergano wrote,
 * "In the [retracted] CSDA study, the researchers compared the normal "entrepreneurial" style of collaboration between top scientists against papers written as collaborations among students of one "mentor" professor. "The authors speculate that the entrepreneurial style leads to peer review abuse. No data is provided to support this argument," Carley says...".

Carley also said she would not have recommended that the paper be published without major revision.

Plagiarism
"Following reports in the Deep Climate blog, USA Today...confirmed with plagiarism experts that parts of the report appear cribbed from the same scientists [the report] criticized, and from Wikipedia." Passages in the background section on paleoclimatology "had been apparently lifted nearly verbatim from Bradley’s Paleoclimatology: Reconstructing Climates of the Quaternary, and then edited in a manner that introduced distortions and errors." The "Wegman Report’s social networks background section...is based on unattributed material from Wikipedia and two widely used text books."

Weakening of Bradley's text
In the Wegman report, Raymond Bradley's original writing "is methodically weakened, silently contradicted or even directly inverted", said John Mashey in his "Strange Falsifications" report. One example of text added to weaken Bradley's account, involves this added passage, inserted between Bradley passages:
 * "In addition, oxides of nitrogen are formed in internal combustion engines that can be deposited as nitrates also contributing to fertilization of plant materials. It is clear that while there are temperature signals in the tree rings, the temperature signals are confounded with many other factors including fertilization effects due to use of fossil fuels."

Mashey notes that this added text is of dubious quality and does not cite a peer-reviewed reference, saying of Wegman's invoking "internal combustion engine" and "fossil fuels" influences,
 * "There is much less evidence for any large-scale effect, especially as most relevant tree-rings are remote from dense human habitations. Internal combustion engines are relatively recent on a millennial scale, so they certainly do not affect most of that time.  This claim is unattributed and likely originated outside the peer-reviewed literature..."

Reference padding
Of the roughly 80 references in the work, half are not cited in the text, and some of these references are bizarre. Reportedly a Barton staffer provided Wegman and his coauthors with the material.

March 2010: unspecified allegation
TBD. "There are two sets of allegations against Dr. Wegman. The first was received in March, 2010..."

April 2010: complaint by Raymond Bradley
On October 8, 2010, USA Today revealed that George Mason University was investigating Wegman Report author and GMU statistics professor Edward Wegman : "The investigation followed a formal complaint by paleoclimatologist Raymond Bradley, co-author of the seminal (and controversial) 1998 and 1999 “hockey stick” temperature reconstructions. ...." John Mashey noted that "Plagiarism [was] only the simplest problem"; of a number of "strange scholarship" issues, "for simplicity Bradley limited his complaint to plagiarism whose verification needed zero field expertise."

October 2011: still in the "investigation" phase
A year later, GMU spokesman Dan Walsch said, "There are two sets of allegations against Dr. Wegman. The first was received in March, 2010, and the second was received in October, 2010. We are referring to the overall process as a "proceeding" ...it is in the investigation phase".

Inquiry timeline
''There's some confusion here about the initial complaint. (Bradley said elsewhere he'd sent letter in April?)''
 * 2010-03-05: Initial complaint from paleoclimatologist Raymond Bradley, in letter to GMU president Alan Merten.
 * 2010-07-28, "[in a] letter to Bradley, GMU vice-president for research Roger Stough said he expected a university committee to complete its investigation of Wegman by the "end of September.""
 * 2010-10, date unknown: second set of allegations received.
 * 2010-10-08: "GMU spokesman Daniel Walsch confirms that the university, located in Fairfax, Va., is now investigating allegations that the Wegman report was partly plagiarized and contains fabrications."
 * 2010-10-11: "George Mason University spokesperson Doug (sic) Walsch has clarified that the complaint against Wegman has moved past the preliminary "inquiry" phase and is now under formal investigation." (superseded by this:)
 * 2011-05-26: "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."
 * 2011-10-04: "it is in the investigation phase", Walsch said of the proceeding.

See also George Mason University's Policy Number 4007, Misconduct In Research and Scholarship, and the expected timeline for such proceedings.

Foot-dragging and lack of transparency from George Mason University
In Dec. 2010, blogger Deep Climate wrote: "More than nine [now 18+ - ed.]long months after [allegedly plagiarized author Raymond] Bradley’s initial formal complaint [to George Mason University ], it is still not known whether GMU has managed to complete its inquiry, or whether there will be a formal investigation. What is clear, however, is that every step of the way GMU has not respected its own procedures and timelines, leading to unconscionable delay and obfuscation. Even worse, it appears that GMU has flagrantly disregarded its obligations with respect to possible misconduct in federally funded work."

Nature: Delay serves no one
A May 26 2011 editorial in Nature concurred that foot-dragging was particularly undesirable in a case such as this one that had influenced policy, saying: "'Long misconduct investigations do not serve anyone, except perhaps university public-relations departments that might hope everyone will have forgotten about a case by the time it wraps up. But in cases such as Wegman's, in which the work in question has been cited in policy debates, there is good reason for haste. Policy informed by rotten research is likely to have its own soft spots. ...'"

Related SourceWatch Articles

 * George Mason University
 * Edward Wegman

External resources

 * The Wegman Report (92 pp pdf) -