A Dummies Guide to Strange Scholarship in the Wegman Report: The Shorter Version
For those bunnies with neither the time or inclimatation to read Eli's Dummies Guide to Strange Scholarship in the Wegman Report (pay attention, you know who you are) Rabett Labs offers the shorter version, together with a few notes on why this is not going away despite the furious spittle projection coming from the direction of Stephen McIntyre and Anthony Watts.
Plagiarizing
- Evidence of extensive plagiarism in at least two sections of the Wegman Report, that describing proxy methods (plagiarized from Ray Bradley's book, see point 1 below for confirmation) and that describing social network analysis.
(See Strange Scholarship in the Wegman Report, the complete handbook, by John Mashey)
- Multiple use of references to support claims that the references contradict. In such cases, while one can disagree with the source, the disagreement must be noted to inform the reader
- The Wegman Panel restricted the social network of climate scientists to Michael Mann's coauthors forcing him to the center of the network. They then mislead the reader by claiming this as evidence of collusive and poor peer review. Incompetence or design?
- Use of a figure from the 1990 IPCC Report altered to support the conclusion of the Wegman Panel, coupled with an unwillingness to consult the original source, even though it was available in multiple places in the Washington area including GMU, as well as from Amazon.
1. We have from the Weg hisself, on his Facebook page, that he was effectively forbidden to supervise graduate students in late AUGUST. For the hard of learning, Wegman has been academically emasculated with a sharp knife.These three points alone are strong evidence, but the chickens fleeing the hen house behavior of Steve McIntyre and Anthony Watts are what clinches the case. McIntyre is following his usual mode of operation, throwing everything against the wall he can find in the hope that something sticks. For those into this stuff a useful watchword would be "Remember Yamal", McIntyre's attempt to smear Keith Briffa. There was nothing there, actually, it was McIntyre who behaved badly, loudly demanding in public records that he already had and claiming that his not getting them from Briffa was evidence of Briffa's misconduct. Oh yeah, his analysis was also dicey but it wasted everyone's time and penetrated much more deeply into the media than the refutation.
2. We have from Donald Rapp's letter that Elsevier is complaining to GMU that Wegman plagiarized Bradley's book. Elsevier has some pretty sharp intellectual property lawyers and would not expose itself to any counterclaim without believing the evidence is clear and convincing.
3. We know from Dan Vergano's report in USA Today and Deep Climate's follow up that George Mason University has concluded that the complaints it has received contain enough evidence to move from the less serious inquiry phase to a formal investigation of research misconduct against Edward Wegman and his group.
Watts, well, he was busy nailing Hal Lewis on Martin Luther's door. Look over there Lucia.
Posts
93 comments:
Expect more leading up to the next climate conference. The 2 main sites are in full flow:
http://wattsupwiththat.com/2010/10/22/mike-manns-secret-meeting-on-the-medieval-warm-period/
the secret meeting (info published on web but hey...!)
http://climateaudit.org/2010/10/15/fiona-fox-and-the-babe-magnet/
Not even sure what this is about but it will see an end to the warmists lies that's for sure. (it's a good job they didn't find out about the tea lady - shes's spoken off record to both Jones and Oxburgh - wow!)
It looks like the kiddie corps has no clue what a scientific meeting announcement looks like.
GMU is indeed taking this seriously by prohibiting Dr. Wegman from supervising graduate students. In fact, several of his students have been accused of plagiarism in their theses. This is indeed serious and is likely to be very problematic for their careers. It's one thing if one student commits plagiarism without the advisor's knowledge. But in this case, it really looks like the students were following their advisor's lead.
I'm still thinking about Hal Lewis, his resignation, and his defenders. What's up with Happer? He's a very distinguished AMO physicist, very smart. His recent activities won't help his reputation. So why is he doing this?
Nevada Ned
Happer is a long time Jason, went to DOE under Bush I and has long held these opinions. He was a purposeful pain in the ass to the Clinton administration, which eventually forced him out of DOE.
As I used to day in happier times in the last century,
Hot Diggity Dog!
Deepfelt thank you to John Mashey & DeepClimate.
thefordprefect and Deech56,
Yeah, I've made a couple of comments on that thread. They really don't know what normal science looks like. Looking at the abstracts, it seems pretty standard. Latest results, open questions, what additional data is needed and possible directions for future research. Real nefarious.
And of course, the fact that it was held in a place where people might want to go, as opposed to say Glendive, MT (where people might actually want to go for a paleontological conference, but that is beside the point). But they ignore the fact that for any conference people have to travel so it really doesn't matter where it is held, in the minds of the wackos they are just jetsetting around and probably traveling by private plane. Yeah, right.
Back in the day, when I worked for companies which actually valued their workers, I was often sent to software engineering/CS conferences here in the US. They were often held in places which were (at least on the surface) fun places to be. Disneyland (in the exotic locale of Anaheim), NOLA, NYC, Chicago, Seattle, SF, DC. None of these places sound exotic to us in the US, but to people from other parts of the world? You tell me, they are all considered vacation destinations (especially Anaheim, NOLA and NYC). As I pointed out over there the main point of these things is always the face to face stuff and a good selection of bars and restaurants (French, Italian, Mexican, Chinese, Thai, Ethiopian? Or maybe something more exotic, like a place where you can eat an alligator before it eats you!) is conducive to the exchange of ideas.
I guess the most amazing thing about the commenters at WUWT is how little they understand about human interaction. Sort of par for the course, of course a nice round of golf on a nice day is a pretty good place for social interaction too. YMMV.
I understand the concern about plagiarism of the DESCRIPTION of "social network analysis."
I do NOT understand the implied acceptance of the technically accurate use of "social network analysis" to scrutinize scientific conclusions. It seems to be a technique perfectly applicable to political assassination as demonstrated by Wegman, et al.
Excuse me for not reading the entire Mashey tome, but the summary of the objection concerning "social network analysis," as given above, is only that it was not applied PROPERLY ... as opposed to it having been employed AT ALL. ("The Wegman Panel restricted the social network of climate scientists to Michael Mann's coauthors forcing him to the center of the network")
Will someone give me an generally accepted example of the technically accurate use of "social network analysis" as an independent measure of scientific validity? Then, please, also offer a generally accepted example of the technically accurate use of "social network analysis" as an independent measure of LACK of scientific validity.
John Puma
Oh my, oh my, oh my. Via SheWonk, as posted by willard, this discussion between Donald Rapp and Bernard J, et al. is proceless:
http://wattsupwiththat.com/2010/10/08/on-wegman-who-will-guard-the-guards-themselves/#comment-505015
for starters
*priceless
John Puma, your point is correct of course, and was made already by Mann et al. in their response to the Congressional hearing, presenting the example of the physics community -- complete with a picture from the Solvay Physics meeting in Brussels 1927, where 29 people met who together defined pretty much what today we call physics. If such collaboration is not acceptable, there is no honest science.
John Puma and David Benson,
I have a problem with the use of SNA in the context of science. First, it does not take into account that individual scientists may carry great influence with a larger group of scientists--those who find their techniques persuasive and useful--whether or not they have published. The Solvay conference is an excellent example. Just about anything Fermi said would carry tremendous weight in part because he was a damned good physicist and in part because he maintained good relations with all and cultivated a reputation for objectivity.
Second, the SNA does not consider the most important aspect of consensus--the evidence and the predictive power of the various aspects of the consensus.
Frankly, I don't think most social scientists have an adequate understanding of these aspects to make an informative SNA.
> supervise graduate students
Is that info still public somewhere?
How does one identify "eligible" faculty from the list?
Wegman's still "faculty" here: http://catalog.gmu.edu/content.php?catoid=15&navoid=1031
Hank,
That little tidbit is found on Wegman's Facebook wall.
"Watts, well, he was busy nailing Hal Lewis on Martin Luther's door."
Horatio was under the impression that Watts is the door, upon which various individuals nail their theses about this that and the other.
Who would be so crazy to agree with Wegman?
Vintage june 2006:
> [M]ore statisticians should be consulted in paleo-climate work. Actually, on this point most people would agree – both fields benefit from examining the different kinds of problems that arise in climate data than in standard statistical problems and coming up with novel solutions, and like most good ideas it has already been thought of.
Source: http://www.realclimate.org/index.php/archives/2006/07/the-missing-piece-at-the-wegman-hearing/
Please also bear in mind that the Rapp discussion was brought to my attention by John Mashey, somewhere in Deep Climate's blog comments.
Ray, sure. I do want to point out in defense of John Mashey (as if he needs any) that, even for a method that is evidently inappropriate as evidence for guilt by association, it does have merit, beside entertainment value, to rub in that Wegman et al even botched its application, in a way that even a dummy bunny may find hilarious... so, Mann was at the center of his network of co-authors. Surprise... so am I.
John Puma
Agreed.
The idea that one can somehow determine the validity of the science (or lack thereof) by looking at who Mike Mann's "friends" are is absurd on its face.
Horatio would guess that even Wegman now sees that inclusion of Said analysis in the initial report was a mistake on several counts.
In fact, in light of how things are playing out (eg, with plagiarism charges), Wegman is probably kicking himself for not restricting his report to statistical arguments (eg, about PCA.).
Martin,
John certainly requires no defense against me--I'm a big fan. His work is an excellent example of what a keen analytic mind can uncover about anti-science.
And really I have no problem with sociology of science, STS or other such studies. Rather, my beef is with those who insist on "doing it wrong". That is, I do not believe it is possible to come up with meaningful analysis of a community--be it an Indian village or a particle physics experimental group--unless you have bothered to understand what motivates the individuals and the group as a whole. The idea that sociologists have found such universally valid methodology that they can conduct analysis while remaining 100% ignorant is precisely what makes them so easily sokaled.
Willard,
Had the Wegman report stuck solely to his area of expertise, it could have furthered discussion and indeed such recommendations as you quote could have been seen as reasonable.
Instead, they turned the report from a commentary on statistical technique into a political weapon. In doing so, they both plagiarized and distorted the science. The plagiarism was a tool to make them sound more credible than they were, and the distortion was required because the evidence doesn't support their conclusions.
This was possible only because you had Congress (and now the Virginia AG) poking its nose where it did not belong. Had Wegman confined his criticism to the peer-reviewed literature, the political diatribe would never have seen the light of day, and he would be spared his current troubles.
1) Read SWSR, W.5.1-W.5.5, and W.5.6.3.
2) Of course in the blogosphere, it is a strange idea to actually ask experts, but in the real world, one does.
I got in contact with a well-published SNA researcher (Garry Robins) who took the time to look at this, gave me comments and OK'd me including them (that's W.5.6.3). A bit later, I checked with a 2nd such researcher, whose comments were similar. I figured the horse was dead and didn't need to waste someone's time beating it again, so I didn't include the second.
3) But, to summarize the problems highlighted in SSWR W.5.
a) SNA people don't limit relationships to coauthorship.
b) Coauthorship analysis, contrary to beliefs of the Wegman team, is not something magically new, but very old.
c) When you start with someone's coauthors, *of course* they have *high centrality." You have to look at the overall network.
d) The general ideas are directly contradicted by other research.
e) Mann's 2006 network is totally irrelevant to his network in 1998/1999.
f) And of course, the WR & SAI2008 leap from nice graphs to claims of wrongdoing, with zero evidence of the latter.
g) And of course, if one reads SWSR, all this is in service of Meme-b, whose likely origin is shown in W.8.9, although it may well have been inspired by Pat Michaels (SSWR p.10, Memes and Themes).
Anyway, all this is covered in SSWR W.5, along with absurdity of someone with Wegman's seniority and huge network attacking that of a (1999 recent post-doc) young researcher, and bypassing peer review to get it into journal in 6 days... and remember, that one Ack'd 3 government agencies for support for research with zero relevance to those agencies' missions part of DHHS, and hence is of immediate interest to ORI. Maybe the Rabett will discourse. I've used the JabberwORI as an analogy, but perhaps for lagomoprhic analogies, "Killer Rabbit of Caerbannog" write large, might fit better.
This (WR 5 and [SAI2008]) is an incompetent mis-use of SNA by people with little familiarity with the field. I am happy to hear *expert* arguments against that.
"Coauthorship analysis, contrary to beliefs of the Wegman team, is not something magically new, but very old."
Indeed so. When I was a student, long ago, we talked about Erdos Numbers.
Nick, I especially enjoyed the "Erdos-Bacon number", where the person has to have both a connection Erdos and a connection to Kevin Bacon.
If Edward Wegman had kept to math
He wouldn't have to take a bath
Instead he jumped in the muddle puddle
And brought about this current trouble.
1) a_r_i_d__s: please do not tar sociologists in general with the problems from mis-use of SNA by statistics grad students and very junior faculty. Social science folks range from very bad to very good. ti is fairly clear that none fo the latter ever saw any of this until recently, and those that have, have expressed clear and very negative opinions. It would have been truly amusing had they submitted [SAI2008] to Elsevier sister journal Social Networks.
2) Erdos: as noted in SSWR, too bad my papers with Kernighan were CS, not math, or I would have had had an Erdos 4.
The ogre in the closet is that Wegman HAD to fill in his panel report with cut and paste, because there was no way he was going to redo the analysis using what he considered to be a proper statistics. He knew what the answer would be, a hockey stick, pretty much the same as MBH. The only way out of that is to do a McIntyre and pick and choose from the data.
Crowley had very early on simply co-added all the proxys to get, wait for it, a hockey stick. So if a brute force method like that works, you know what the fancy stuff will get you on the global record (the fancy stuff does get you geographical distributions).
Horatio said he "was under the impression that Watts is the door".
I'd always thought of aW (attowatts) as a bit of a plank. Now it seems I underestimated him. Being a door, he has to be at least 5 or 6 planks.
Cymraeg llygoden
Cymraeg llygoden --- Just a wide dense plank.
Ray,
Considering what you just said, I believe you might agree with what Dr. Hans von Storch has said "under oath" regarding "sharing" code:
> [T]he documentation must take the form of a mathematical description of the algorithms used. This is in many if not most cases sufficient for replication. Also, the intention of replicability is not to exactly redo somebody’s simulation and analysis, but to find the same result with a similar code and different but statistical equivalent samples. The problem is usually not that the codes contain errors (even if many of the more complex ones likely contain minor, mostly insignificant errors), but that specific elements of implementation and specific aspects of the considered sample of evidence will lead to conclusions, which do not hold if another sample is considered or a different but equally good algorithm is employed. The reason is that we want to learn about the dynamics of the real world, and these insights should not depend on random choices in sampling and implementation. We generally do not expect scientists to manufacture results, or that unintended but significant errors will affect the eventually published conclusions. Having this situation in mind, I consider Rep. Barton’s requests to the three scientists as inadequate and out-of-scale. However, the language used by Rep. Barton makes me perceiving this request as aggressive and on the verge of threatening.
Source: http://frwebgate.access.gpo.go.....31362.wais
Horatio is uncertain about many things, but he's pretty sure that Anthony Watts is no planck (not even a particularly dense one).
And certainly not 5 or 6...
John Mashey,
My intent was not to paint all sociologists, or even sociologists studying science, as charlatans. However, those who merely apply their "method" without understanding either the subject matter the scientists are studying or the culture of the subfield are merely postmodern idjits. I say exactly the same thing about anthropologists studying "primitive" tribes without understanding their culture. Informed criticism is important, even welcome. Ignorance masquerading as method is self-parodying (ala Sokal).
> I say exactly the same thing about anthropologists studying "primitive" tribes without understanding their culture.
Here is an example of the "anthropologist" standpoint:
> I often mention how I feel like an anthropologist in seeing how little scientists understand about the process of a prospectus [...]
Source: An interesting discussion between Steve and John Hunter, around the notion of conflict of interest.
a_ray_in_dilbert_space (7:35 PM), is this anything like the "we're from the statistics department, and we're here to help" syndrome evidenced in McShayne and Wyner or especially the WR report?
I've come to the definite conclusion that statisticians in isolation, with no climate scientists to advise or call upon, are a pointless exercise in futility when analysing climate data. Just as well that there are statisticians who specialise in climate and do talk to other scientists. A shame congressmen, senators and members of Parliament decide to either ignore them or haven't even heard that they exist.
http://www.caitlin-buck.staff.shef.ac.uk/SUPRAnet/
http://www.caitlin-buck.staff.shef.ac.uk/SUPRAnet/programmesummary.pdf
"SUPRA-net is a statistician-led, international, interdisciplinary team focused on such uncertainties."
Or do they pretend that they don't exist?
"The Wegman Panel restricted the social network of climate scientists to Michael Mann's coauthors forcing him to the center of the network. They then mislead the reader by claiming this as evidence of collusive and poor peer review "
"Short-centering" to get the result they wanted?
Willard,
It has been my opinion that when an anthropologist reaches conclusions that do not provide insight to the people studied into their own culture, or which are unrecognizable to the people being studied, it is invariably the anthropologist (the outsider) who is wrong.
Nevertheless, many social scientists resist learning about the culture they are studying from the members of that culture. I have actually heard social scientists claim that they are trying to avoid bias. I would contend that what they are trying to avoid is work and having to alter their own preconceived notions. Stevie would be welcome to learn about the culture of science, but to do that he'd have to publish in real science journals...and that would be work.
Ray,
Here is another example:
> I’ve often described myself as feeling like an anthropologist in studying the behavior of climate scientists, because their standards of replication and audit (or lack of them) seem as foreign to me as tribal customs must have seemed to early 20th century anthropologists in the South Sea Islands.
Please note that the thread has 270 comments. (Deech might recall it.) This should provide evidence of some kerfuffle among specialists of the Scientific Method, all being for Science Betterment. Someday, I might trademark that expression.
I note this comment from Steve:
> Personally, I think that it is very unlikely that there are fatal errors in all the underlying studies being relied upon by IPCC. I am far less dogmatic than many readers.
It might be important to recall from time to time that Steve is no skeptic.
Willard,
A denialist is not someone who denies there is a greenhouse effect. That would be more accurately termed a nut-job or a crank.
Rather, a denialist is anyone who denies or otherwise refuses to consider the evidence. Steve has focused obsessively on only a tiny fraction of the evidence that is out there, so while not a denialist, I would not call him an objective voice. He is also perhaps one of the great nonpublishers in scientific history, being content to blog about the work of others--and hide behind his more aggressive commenters--rather than doing actual science himself. He has done nothing to advance understanding of Earth's climate.
Wow, willard, a thread blast from the past. I could add so much more now... (and once I did a search and found that someone had referred to my post in a different site - I felt special ;))
CA always did put up with me. Anyway, my arguments in that thread just added more fuel to my general point that it's hard for non-scientists to write authoritatively about the conduct of science, whether it's about publications or meetings. I have been trying to make the point in various venues that climatologists conduct science in the same way that other scientists do.
I do think a comparison of the complexities in biological systems and climate systems would be fun to discuss.
Deech56,
When it comes to comparing complexity in any physical science to that in biology or ecology, I simply step back in awe. And yet, science is sufficiently flexible to make progress even in the biological/ecological domain, where nothing may ever be truly repeatable.
Physical scientists are such arrogant SOBs because what we study is so simple.
Here's a good article on denialism.
http://eurpub.oxfordjournals.org/content/19/1/2.full
They define it as ” the employment of rhetorical arguments to give the appearance of legitimate debate where there is none, an approach that has the ultimate goal of rejecting a proposition on which a scientific consensus exists."
I'm certainly in favor of Scientific Buttermint.
Sounds delicious.
And here is lurid proof pf the bipartisan indictment of the report :
http://reason.com/blog/2010/10/25/what-should-you-wear-to-the-go
Deech "I do think a comparison of the complexities in biological systems and climate systems would be fun to discuss."
Yeah, I raised this yesterday with someone. There's a newish idea around that obesity is the body's protective mechanism against a bad diet - and that the consequences (diabetes, heart problems, etc) arise when that sequestration process breaks down or is overloaded in some way. Very interesting.
But it doesn't change the general view that avoiding obesity is a very, very good idea. The details of the biological mechanisms are (almost) irrelevant for people and their general practitioners and their decisions about diet and health generally.
MinniesMum
Oh OK. Forgot the important bit.
The physical climate processes and the subtleties of measuring and modelling ever finer detail are a good parallel. Just as we can't predict whether an obese person's knees will be damaged before any long-term impact on their heart. We may not be able to say with any certainty when or how bad droughts or floods or other impacts might arise with the climate. We do know that all those impacts are likely and it's a very, very good idea to avoid them.
MinniesMum
MinniesMum, you raise the exact point I was trying to make.
I was trained as an immunologist and worked for a while as a vaccine researcher. The immune system is incredibly complex and there is so much we do not know, including how each person will respond to a given stimulus. Despite of all that the uncertainty, we know enough to say that vaccines work (never mind the cranks who say vaccines cause autism).
Failing to act because of uncertainty is, as we used to say long ago, a cop-out.
My, my, my. I arrived home yesterday, and what should I find awaiting me? Why it was a missive from the lying sacks of rat feces at the Cato Institute asking me to join them in their fight against physical and economic reality. I suspect I must have words with the subscription department of The Economist, as that is the only possible way I can see that they could wound up barking up such a horribly wrong tree.
Cato must be doing well. Nice paper, and over 5 pages, double-sided. I can recycle the paper, but I'm wondering how to handle the request. In an ideal world, I and people like me could get them bankrupting themselves in a pursuit to win over our hearts until they collapse and die of a broken heart. Then we could piss on their graves. I had thought of writing them a little reply saying that having to predicate their whole philosophical outlook on lying about science isn't a really good way to win over a scientist. I'd also thought about pretending to be interested, but sending them a quarter and saying it's all I can afford in such a tough economy. That way I might stay on their mailing list and hopefully further drain their funds. Any creative ideas?
> Personally, I think that it is very unlikely that there are fatal errors in all the underlying studies being relied upon by IPCC. I am far less dogmatic than many readers.
"Some of my best friends were Jews..."
This is soo fake. Just because he denies only the small corner of the science that he made his hobby (and which isn't even all that important), doesn't make him any less a denialist. Just like a racist hating only blacks and having no beef with native Americans, Asians or Latinos, is still a racist. Just not equal-opportunity.
Eli wrote:
Crowley had very early on simply co-added all the proxys to get, wait for it, a hockey stick. So if a brute force method like that works, you know what the fancy stuff will get you on the global record (the fancy stuff does get you geographical distributions).
This is well worth a post spelling it out for some bunnies.
Pete Dunkelberg
re: Wegman and Facebook
Eli mentioned this, but left out some of the good stuff.
1) I first saw this over at Deep Climate via Derecho64, and then Andy S W quotes:
"The comment on his Wall from August 21 painful to read:
Edward J. Wegman Want to know a bad week? All in the same week. 1) accused of plagarism, felony, anti-science, misleading Congress because of your climate science testimony, 2) have a rule made up, which only applied to you, that blocks you from mentoring graduate students, 3) have a friend tell you he was not happy with you because you were awarded a patent.
August 21 at 4:17pm"
That is the week during which files suddenly disappeared: Wegman's C.V., Said's dissertation(2005), and her key 2007 seminar [SAI2007]. The last also got edited out of the GMU seminar record, although not elsewhere. See SSWR A.11 for details. See especially A.11.2, the annotated copy of here talk, illustrating why it was probably a very bad idea in the first place.
2) For amusement, further back in the wall was:
"Edward J. Wegman
So here we are with 30 inches of snow measured in my yard. It is up to some of our windows on the first floor. As Anan says, Global Warming. I guess this snow is due to global warming!
February 6 at 5:52am”
Well, do recall that Wegman had trouble believing Greenhouse Effect in higher troposphere, given that CO2 was heavy. (SSWR A.2).
He may also be unfamiliar with FaceBook privacy controls. Now privacy and security are not the same, but he has certainly written some Internet security-related papers....
Lol, this blog is still around?
--No One
The evisceration of M&M and Wegman continues at DC's place.
http://deepclimate.org/2010/10/25/the-wegman-report-sees-red-noise/
MapleLeaf
"The comment on his Wall from August 21 painful to read:
Want to know a bad week? .... 3) have a friend tell you he was not happy with you because you were awarded a patent."
You just gotta wonder "why?
Though it may not apply in this case, of course, Horatio can think of one possible reason: the "friend" felt he/she was actually the originator of the "invention" OR at least should have been included as a co-inventor on the patent for significant contributions thereto.
Horatio has experienced this first hand -- and is still "unhappy" about it years later.
Maybe even as "unhappy" as one might be if one has had one's words and ideas "lifted" from one's book and extensively reproduced without permission or even attribution...
Maple Leaf says, "The evisceration of M&M and Wegman continues at DC's place."
Really, evisceration is such an ugly word--and it presumes thay had any guts to begin with.
Maybe "defenestration" is a better term?
> Maybe "defenestration" is a better term?
Nah... that's when I help people upgrade to Ubuntu ;-)
What about "Crime Scene Investigation"?
Hmmm... word verification offers 'gonato'. "Degonadization?" (ducks and runs)
The evisceration of M&M and Wegman continues at DC's place.
http://deepclimate.org/2010/10/25/the-wegman-report-sees-red-noise/
That looks more like "emathsculation" to Horatio.
Hi Rattus et al.,
Yes, "evisceration" is not a nice word, nor would I wish it on anyone, of course. I have not heard of "defenestration" in a long time; throwing someone out the window is not a good thing either. Then again, way back when, they used to throw the dirty dish water out the window, so maybe that description is indeed appropriate as it applies to McI et al. and their statistical skills ;)
How about "demolition", "obliteration" or "annihilation"? I also like Horatio's "emathsculation".
MapleLeaf
PS: Now I wonder if my Mac is going to start acting up after posting this.....could it have something to do with folks affiliated with M&M?
Horatio,
Win!
After reading the exchange between Bernard J. and Donald Rapp on WTF, I'm struck again by how creepy and stupid the obsession with pseudonymity (which they continually and erroneously refer to as anonymity) is in a venue where pseudonyms are less rare than real names and also where real names are pretty much unverifiable.
*cough* Steve Goddard *cough*
What makes it even worse is that while they're puffing out their chests on their virtual high horses, they're guilty of a sin much the same but actually worse. They're insisting on having a voice equal to experts, without having any expertise. The name issue is just a name. The expertise issue is real.
Emathsculation
Note an odd inconsistency:
1) Letter from GMU 07/28/10 was the first time they told Bradley they had formed an inquiry committee in April.
2) GMU policy" says they tell the respondent when they form an inquiry committee, and they get to challenge.
3) But Wegman's Facebook entry 08/21, and the sudden disappearance of his C.V. and other files that week, seem to indicate either:
a) He had been told in April, but hadn't taken it seriously OR
b) He hadn't been told in April OR
c) GMU really had not formed the committee in April.
There is a certain Alice in Wonderland quality to all this.
Snow Bunny says:
I'm willing to think Social Network Analysis is valid in the right hands. Applying it to climate science is an utter crock. Climate science, as a whole, is has almost as many axons as a neuron in the brain. The theory requires input from many specialities in physics and chemistry, and the data showing global warming agrees across a wide variety of fields.
For instance, the biologists' data show earlier springs and later falls, species moving uphill and northward, etc. In agreement with predictions. The gravity satellites GRACE discovered more mass is being lost than the glacialogists had expected. Fields where people weren't specifically looking for global-warming confirming data.
The notion of climate science as propagated by a small claque is utter bunk. Wegman's group must have grossly misapplied it.
Snow Bunny says:
The rise in temperature is so marked on the curve it doesn't need statistics. The target is so big a blunderbuss will hit it. You don't need multivariate statistics. It's a simple two-dimensional plot, temperature vs. year. You can pick it out with the naked eye.
The idea that Mann, or anybody, could have used the wrong statistics is ludicrous.
I've hitherto presumed that the "April" committee was the initial inquiry committee that were expected to report by September on whether there was sufficient evidence to form a formal investigative committee.
So, was it in mid August that the initial inquiry committee, i.e. the "April" committeee, deemed there to be sufficient evidence that required a formal investigative committee (the "August", or august, committee) be set up and it was then that the Wegman documents "disappeared" and the FB wall comments occurred?
If that is the case, then the inconsistency might disappear mightn't it?
That said, things seem to be dragging a bit, which can't be good for Prof. Wegman's reputation if it is found by the "August" committee that Prof. Wegman has no substantive case to answer and that his reputation must be restored.
On the face of it I can't imagine that being the outcome, but just imagine what the attowatts and McFly-by-night blogs will make of that if that transpires.
I bet their strategies are in place whatever the outcome.
Cymraeg llygoden
Cymraeg:
See comment at DC, which gives the timeline in nominal elapsed days according to GMU policy.
The *inquiry* committee was (supposedly) appointed in April. if this takes more than one short meeting to examine and say "Yes, need investigation", then something is wrong.
I will check your timeline tomorrow John, but there is also this in the GMU rules:
"The inquiry committee completes the inquiry, including the preparation of a final inquiry report that includes any comments received from the respondent, within 60 days of the committee’s first meeting unless the Dean or Director determines, and documents in the inquiry record, that the circumstances warrant a longer period."
So, while one would hope it would be much shorter than 60 days, it could conceivably have taken, what(?), until the end of June just for the inquiry committee to say yes there's a case to answer according to its own rules.
And from GMU's own letter you link to it seems they were having problems getting people to meet up then.
So was that the Dean/Director saying then that the initial process would overrun the 60 days? Seems likely, now.
And yes, I'd agree that's probably not good enough. But the wheels of due process do grind slow at times, especially when lawyers get involved too.
Cymraeg llygoden
Cymraeg:
1) An allegation of serious plagiarism, including a senior professor, where the material is already public is something most schools would want to handle as fast as possible, either to definitively clear someone or get an investigation underway.
2) Let me set up a thought question for people:
Here is the nominal timeline, as at DC, i.e., the first number is the cumulative elapsed days, the second is the nominal interval for that activity. The letter codes correspond to annotated copy of the policies I have.
Nominal GMU timeline, approximate, given
Elapse Interval (as soon as possible) everywhere, challenges.
0 0 A Allegation
14 14 B See if inquiry warranted
28 14 C If so Provost appoints committee Respondent may challenge
28 D First meeting of inquiry committee
88 60 E Inq. Com. completes report. Investigate? (Y/N)
102 14 F Dean/Director determines. Investigate? (Y/N)
132 30 G VP convenes investigation committee
132 Respondent may challenge
132 Respondent comments, etc.
252 120 H Invest. Comm reports, best efforts 120 days
252 VP reviews report, sees if university accepts
282 30 I Possible appeal
382 100 J President writes decision on appeal
So, here is the thought question:
Suppose Day 0 (A) had been ~ 03/15/10 (Monday)
Suppose the first meeting of an inquiry committee (D) had been been Friday 03/19/10 (+4 days), an inquiry completed by 03/24/10. That would be amazingly fast, but would certainly show the university recognized the issue and acted quickly. That's one imaginable extreme. The nominal timeline above would predict no later than +28 days.
Now, I'm curious to hear people's opinions, especially anyone with relevant academic experience. Obviously, this is subjective.
By what date (D1) would people start thinking GMU were pretty slow?
By what date (D2) would people think GMU was doing serious foot-dragging, maybe even stonewalling?
Remember, we are just talking about the first meeting of the *inquiry* committee.
Time for another anthropology comment:
> As someone who’s spent most of his life unaware of academic peer review politics – but familiar with forms of due diligence in other fields – I sometimes feel like an anthropologist among South Sea Islanders in the 19th century.
This is only a rerun. What's more interesting is the comment to which it's an answer. NW's story made me smile. Here it is:
> [A] certain professor (who will remain nameless) became very, uh, skeptical of peer review in midlife. So when he finished a paper, he made a list of journals, in order of his preference, where he thought the paper might fly, and went down the list. If the first journal rejected it, he would: (i) throw away the referee reports unread; (ii) cross off the journal; (iii) wait two weeks; and (iv) send paper to next journal on the list.
Don't forget the now famous **How To Publish A Scientific Comment in 123 Easy Steps**!
John: be careful of overstating your case (not saying you are, just try to think that way).
Maybe you could look at the time, taken for other academic inquiries. for instance the SUNY professor that tha English guy went after (within the climate wars) or perhaps a selection of other cases in academia. Are you really an expert on "how long unis take to investigate stuff"? Or are you just opining on this one case based on a gut instinct? It's at least possible that slow turning wheels are the norm. Not justifying that either. but it would make it a general problem rather than a variation from the norm.
Anon:
?? I wasn't stating an opinion or gut feel in the previous post, but just gave the nominal timeline and solicited opinions from others who might be familiar with such, as such information is nontrivial to get unless one is directly involved or has some reason to know about it.
However, experienced academics would be able to make general comments about how this usually works in their schools.
Unfortunately, John, such an academic would have to be very "experienced" indeed. These cases are rare enough that many people won't deal with a single one in the course of their careers.
I think the Wegman case breaks new ground. The accusations and evidence are highly visible and accessible but not "published" in the sense of being reported via established academic channels. I suspect the GMU admins take (false) comfort from the belief that it's no big deal for their school's reputation because it's all "only" happening on blogs. The "people that matter" to the GMU admins (senior admins and faculty at schools GMU wants to impress) may or may not have noticed yet---I don't know. The belief by GMU that other people haven't noticed is the only explanation I can see for GMU's dilatory handling of the case. Can anyone point to on-the-record acknowledgment of the issue by senior people outside GMU?
Ted:
1) I do know some folks who've handled a few.
2) There's nothing on the record I know of, BUT:
offhand, I know of 6 quite credible universities where senior people (mostly Dean or above, or at least full professors heavily involved with academic integrity issues) are watching this, and many of those schools are non-obvious, which leads me to expect there may be more. There are a bunch more where I don't know that, but know that various professors are interested, and not just from climate science.
3) Of course, if you check Table A.6.1 in CCC, you might want to look especially at A.6.1(b), in which funding of GMU by folks like the Kochs and Richard Mellon Scaife appears. Note that the other underlined entities in A.6.1 are GMU-related. All this may be coincidental or it may be a reason for handling this as they have.
4) Note also that apart from other universities, there is DHHS ORI, with which our lagomoprh host is familiar.
John, are you referring to this think tank?
Why did the fourth person drop out of working on the Wegman report? That's the burning question, but one which is also easily explained by a number of everyday scenarios.
John said:
"4) Note also that apart from other universities, there is DHHS ORI, with which our lagomoprh host is familiar. "
Eli said: Well, not in the carnal sense.
Neven:
As per CCC A.6.1, Mercatus is only one of the GMU entities:
CMPA - Center for media and Public Affairs
InstHumn - Institute for Humane Studies [Singer was there]
STATS
So, with GMU itself, that's 5 separate columns in A.6.1.
Those are all described in CCC A.3.
John, Good to know there's some senior people watching this. I've been disappointed with how many academics of my acquaintance are generally unaware (and uninterested) in the whole assault on science. They don't dispute the need for action to reduce the impact of GHGs and they are often taking measures to reduce their personal contribution. They're just not paying attention to the dynamics of the push back. By assuming it's not their problem they increase the likelihood of it becoming their problem some day if they take a controversial stand in their own field ...
I also think it would help to have senior people put some attention on the editorial processes that led to CSDA publishing Said et al.'s article. The article's "striking similarities" are a real issue but a less important one. Much more important is whether the system was gamed to get such a weak article through review. If important people are asking that question behind the scenes, we are more likely to get an accurate answer.
Thanks, John. I'm reading it now. I can't believe you put so much effort in this. When I grow up I want to be just like you.
I am confident that GMU will do a full and complete job of investigating this. The people at the university are serious scientists who do realize the gravity of this issue. Mason does have a reputation of having some right wing people on its governing board, but that has nothing to do with his procedures, operations or in particular, its faculty. This is not going to be whitewashed or ignored. That simply can't happen.
Having said that, the process is likely to be careful and deliberate. I am quite sure that lawyers will get involved on all sides at some point, and you need to document and follow procedures carefully. I wouldn't be surprised if they are already involved.
However, GMU is also in a no-win situation. If they take strong sanction against Wegman, the right wing will decry this as persecution of someone who went against the "massive climate conspiracy" (or some equivalent nonsense). If they don't take strong sanctions, the climate people will decry Mason as a "bed of conservative climate deniers" by everyone.
For me, this seems like a greek tragedy. I believe Wegman is a serious scientist who wrote a bad paper. This paper was co-opted by a political cause, and convinced it was extremely important and needed to come to the attention of congress. As more people became critical, more time was spent defending the original paper. Then these mistakes were turned into a very sloppy congressional report that was largely ignored by the scientific community. It is likely that sections of it were written by a current or former grad student and it wasn't carefully checked before it was published because of unknown reasons - most likely arrogance and misplaced trust. This probably isn't the first problems in in his career, but lines have been crossed and there are a lot of people who now are focusing on him with the intent of taking him down.
Because of this arrogance, he might lose his job. I am not saying that shouldn't happen, but it is a huge fall for a well known statistician.
Anonymous Ralph
As it happens, GMU is now 7+ months into this without having given a clear indication this has gotten beyond inquiry into investigation, the note in USA Today notwithstanding. Stough claimed in the letter at USA Today that the inquiry committee had been formed in April ... and their rules require the respondent to be informed *then*. Wegman's seeming Facebook surprise in mid-August is fascinating. Either Wegman knew about it and didn't take it seriously, or GMU didn't tell him, or perhaps were doing their best to delay.
For reasons I cannot explain at this time, I am very glad to not to have Roger Stough's job at this time.
A school could decide to do inquiry, form a committee, do one as simple as this, clear someone or recommend an investigation ... in a week an half. [Yes, this does happen that fast.] Schools do much more complex full investigations in less months than this.
Serious people at other schools are following this closely, at the university President, VP Research level ... so are some serious statisticians. Ralph didn't mention DHHS ORI, possessor of serious teeth, not just for individual researchers but for whether or not organizations are doing their job. Read about "debarment" at DHHS ORI.
When an allegation of serious, high-profile academic misconduct arrives, from a distinguished professor at a respected school, but with details *already* public:
a) The school owes the professor a *rapid* inquiry if they deserve to be cleared, especially when the material is already public, but ambiguous. {In this case, there was no ambiguity in the plagiarism, one can argue about which items were fabrication (which is why Bradley avoided that), but the ambiguity of a 3-author report is: who was responsible?
b) The school owes a whole lot of people a rapid inquiry and investigation if it turns out there is a problem. if it becomes clear that a school doesn't handle the *simplest* academic misconduct well, they are headed for serious trouble, not just in academe, but with ORI.
c) So far, GMU has not performed well, and I believe additional information will become available in the not-too-distant future that will back this statement. (So, don't take me on faith, but bunnies should keep eyes open.)
Actually, what's amazing is that Ray Bradley was so collegial in keeping the complaint non-public for almost 7 months in the face of B.S.
It's a bit of a stretch to paint Wegman as an innocent in the forest. Both the report and his testimony show this to be wrong.
GMUs problem is that they have heavily featured Wegman, Wegman's students and Wegman's program as part of their push into research and graduate study. To find him involved in a scandal of his own devising is devastating.
Wegman should just rewrite the report, add citations and direct quotations and sentences like "follows closely from Bradley". And also issue an apology. People could still rip at him for the plagiarism itself and for the report's mathematics. Or say "it's not enough". But it would blow over. And it would not cost him his job. Plus it would be the right thing to do.
The defensiveness is going to get him into more trouble than anything else. I was pretty bugged by the refusal to show his methods to Ritson as well.
He did a sloppy job on a complicated issue. Like McShane and Whyner. Maybe he will learn his lesson that it is not so simple to add value cross-field. That he needs to be more buttoned up when wandering into battle in foreign lands. I mean Jolliffe was very honest about how much effort it would take to truly untangle Mann and MnM and "what matters": full code, weeks, and acess to ask questions. And that was a PCA expert!
I think the Mashey-DC stuff with conspiracies (even the name "deep climate") is overblown. Heck, I could blather about Fenton and CRU if I want, but I just have too much balls for that. That said, Wegman has not impressed on several aspects of scholarship...just hasn't lived up to the billing.
And McI and the skeptics have not impressed with their tauting of him. WAy too ready and way too convenient. ARgument by authority when it helps you, but not when it doesn't. Hypocritical and rinky dink. Why oh why do my skeptics overreach? Why can't they be skeptical of both sides?
Eli: can you say more on "featured"?
"Actually, what's amazing is that Ray Bradley was so collegial in keeping the complaint non-public for almost 7 months in the face of B.S."
That may well be because Elsevier (and/or their lawyers) probably told RB to "keep schtum" as much as any (strained?) collegiality on RB's part.
Cymraeg llygoden
Even I (with my more haste less speed outlook on this, and my innocent until proved guilty belief, since prima facie evidence doesn't always equate to a guilty verdict) think things are dragging a tad now.
Is the investigation perhaps complete, but the reports are being pored over by legal types? Is that feasible without comment from GMU?
Cymraeg llygoden
"That may well be...." is quite plausible, just wrong.
Academe tends to give an institution a chance to show good faith until proven otherwise, because most do, especially on something so straightforward.
Talk to academic friends, some views differ by country.
I am not certain, but I suspect that Mason and most Universities very very rarely have to deal with this kind of crap. I am not surprised at all that this is taking a long time, although I also don't understand the delay between the initial formation of the committee and the other events. However, I think it is extremely likely that having lawyers in the midst of this changes everything.
I also can't imagine the results of this going public in a big way in the form of Mason press release. It would be in the interest of Mason to come to some kind of settlement that doesn't make the Washington Post. Quietly buying someone to get out would seem more likely, particularly if lawyers are taking about suing.
I think releasing the update that Mason on the investigation status by Ray Bradley was a a good choice. Despite the fact that being careful is critical here, the reality of this affecting the reputation of the school. Beyond that... it is good they can't just ignore it and hope it goes away.
Although I don't know exactly what John was referring to, I am also very happy I don't have Stough's job now either.
I would also state that it seems very likely that Wegman's department had no clue that this was going on. Apparently he is "Computational and Data Sciences", and it has a group of people working in simulation and data related fields. None of them are climate researchers. I met the department chair, and there is no way he would tolerate this if he had even the slightest hint there was an issue. I am not sure how heavily he has been featured. I haven't seen him getting promoted by the university on this work.
Anonymous Ralph
Ralph:
1) I know someone who has handled 50 cases at their university, although I suspect most were students. Admittedly, academics of my acquaintance have expressed amazement at this one.
2) The problem with the committee is: it was formed in April, but when did it first meet? I don't want to scoop somebody, but "April or May" isn't the right answer.
3) Re: Stough: the issue is that other universities have an eye on this. Universities must be seen to handle academic misconduct complaints fairly for the respondents and complainants both, and that includes reasonable promptness. Finally, there is DHHS ORI. GMU does get ~$80M/year in Federal grants...
4) Bradley gave GMU every chance to handle this straightforwardly and quietly and they didn't.
John - I agree with most of what you are saying.
There is the scrutiny of this process right now because this was not reasonably prompt. I would hope that Mason - or most schools - hasn't seen this kind of problem before.
I am sure Mason get lots of cases of students doing this on papers. I am an academic, and I have far too much experience with this myself. Sadly, I discovered another problem paper from a student this week. I probably have turned in about 10 students over my career. There are procedures in place, and you follow them. (Usually you turn it over and then follow the recommendation of an honor court.) Failing grade, suspension, or expulsion, depending on the seriousness of the case.
You can't compare that with faculty being involved. That HAS to be a rare event. None of the
penalties or procedures are the same. Students don't have lawyers involved (by university code). The media doesn't care, and there are not political implications of what you do,
I don't believe that institutional disbarment is a real threat. Yes... DHHS ORI has that power. However, to wade into this case when it involves congressional testimony about climate change right after republican's just won control of the house would be agency suicide. Penalizing an entire university of scientists who have nothing to do with this case just won't happen. Imagine the video of the fired graduate students and postdocs giving interviews on Fox news, CNN, and MSNBC about how they lost their job because an investigation went too slowly.
I certainly could imagine individuals facing this, but not the entire university. You don't burn down the forest because there are a couple bad trees.
BTW - I know of some institutions that run all PhD dissertations through "turnitin.com" to check for plagiarism. It is incredible how often these schools find problems in the final drafts. Fortunately, most schools believe their students don't plagiarize, so they don't check their dissertations for plagiarism. This solve the problem nicely.
Anonymous Ralph
Ralph:
1) Yes, we agree, i.e., i.e., I interpreted your "this kind of crap" as plagiarism in general, but noted I expected it was mostly of students. When I was a a computer science instructor 40 years ago, I must have caught ~10 students/term copying final project programs.
2) ORI: we agree. I was in a hurry and should have said more.
According to ORI, they have never debarred an institution, and indeed, I doubt it would happen to GMU. I am just pointing out that ORI has a bigger club than many in the blogosphere realize and that their interest may well grow well beyond Wegman & Said into parts of GMU's research administration. The provost had interesting comments in 2001.
There have been hints of other problems, as well as the potential mis-use of funds, 3 dissertations with plagiarism and unimpressive supervision of PhDs. For example, how carefully is anyone watching when a dissertation cites less than half its references? Only some of this is of interest to ORI, but one of the lessons in this whole exercise is that sometimes many more problems appear when you start looking hard.
The question is whether this is localized or part of a wider problem. I certainly don't know.
Deepending on what really went on, some higher heads might have roll, although there is plenty of murkiness to go around. Given that some well-funded entities of GMU seem to exist to ~eliminate the Federal government, it may well be that funding agencies take a much closer look. I know there are perfectly-reasonable people at GMU (I even sent a few heads-ups so they would be warned).
Of course we also know GMU graduates Cuccinelli and Russell are forcing U VA to waste a lot of money, so taking a hard look at GMU does not seem unfair.
Anyway, I certainly don't think institutional debarment is at all likely, but in talking with academic friends, the idea of messing with ORI just induces shudders, because ORI does have the club, which of course means they don't have to use it.
SOME THOUGHT QUESTIONS: OPINIONS FROM ACADEMICS?
From Dan Vergano's article one finds the teh 07/28 letter from GMU to Bradley.
From that letter, what might you infer about the date of the 1st inquiry committee meeting?
Put another way, at your school, assuming a complaint like this arrived to a VP Research,say 03/15/10:
A: when would you consider a normal interval until the 1st meeting of an inquiry committee?
B: By what date would you worry that something was wrong?
[I am keen to get opinions from additional academics, most helpfully in the next few days.]
Post a Comment