With a tip of the hat to: "Close Encounters of the Third Kind" Image: Alamo Drafthouse Rolling Roadshow - click
Guest post by Thomas Fuller
In the past few months we have seen a number of amateurish attempts to counter skeptical arguments that gained traction in what public space there is for matters climatic and anti-climactic. Today we get introduced to John Mashey’s attempt to smear Edward Wegman and reclaim the Hockey Stick for further usage. It’s definitely anti-climactic.
Climategate had a huge impact on public opinion regarding the probity of some of the scientists involved. The leaked emails clearly showed bad and bullying behaviour that left a stench in any honest reader’s nostrils. Andrew Montford, among others, chronicled Climategate and the events leading up to it in a clear, detailed narrative called The Hockey Stick Illusion. It has been praised by reviewers, including climate scientist Judith Curry. I have read it. It is good. It is accurate. I recommend it without reservation.
So, a few months ago a website called Scholars and Rogues published an incredibly lame attempt by Brian Angliss to show why nobody needed to read The Hockey Stick Illusion, citing the low number of emails that were leaked as evidence that we didn’t have enough evidence. When Steve Mosher pointed out that a crooked accountant probably had numerous accurate transactions to his credit, but that only one was needed to prove him criminal, Mr. Angliss and Scholars and Rogues sort of went away.
The late Stephen Schneider and an IT administrator named James Prall published a paper in the Proceedings of the National Academy of Sciences claiming to prove that scientists on their side of the fence were more credible than skeptical scientists, because they had more publications. Sadly, they only searched in English, they got the names, job titles and specializations of the scientists wrong, they used incorrect analysis techniques (as explained quite brilliantly by RomanM on a post at Real Climate), and used Google Scholar, a commercial database with no published quality control measures, as opposed to any one of several available academic databases. They didn’t get Stephen Schneider’s publications accurate. But that’s okay, they have the names and pictures of those they labeled (incorrectly, in many cases) as skeptics on Prall’s website.
Now comes John Mashey, intent on the destruction of Edward Wegman’s criticism of Michael Mann and Raymond Bradley’s carefully concocted Hockey Stick Chart. Wegman was asked by a congressional committee to investigate their work. His report, fully supported by the National Academy of Sciences, was devastating, citing quite correctly the fact that random noise fed into Mann’s analysis scheme could produce a hockey stick, that they used incorrect analysis of principal components in their study, and that the community of scientists involved was so closely interlinked as to be best described as incestuous, making claims of independent verification a mockery.
Before I go any further, I should note that Mashey makes another accusation that hasn’t been picked up by the media: He accuses Steve McIntyre and Ross McKitrick of being ‘recruited, coached and promoted’ by the George Marshall Institute.
Page 30 of John Mashey's full report
I don’t think that’s true. Mr. Mashey does not provide any documentation for this accusation and I’d like to see his evidence. I certainly hope it’s of better quality than the rest of what turns out to be drivel.
John Mashey’s first bone to pick with the Wegman report is that Congressman Barton’s staff provided source material to Wegman. It is the first item in Mashey’s report, (PDF – warning large download) a 250 page diatribe.
But it is entirely normal that Wegman would ask for and Barton’s staff would provide, any relevant material to speed up the investigation. I have done investigative work for several government bodies and it is in the interests of saving the public’s time and money that papers are provided. I literally cannot understand why Mashey would make this his first point.
Also on the first page of Mashey’s report is the ‘accusation’ that one of Wegman’s associates in the investigation was a post doctoral student with one year of experience, Yasmin Said. Perhaps Mr. Mashey should take a quick look at how much experience Michael Mann had when he created the Hockey Stick that became the iconic representation of climate change to the world…
John Mashey says that Wegman plagiarised material in his report to Congress.
This is odd. Wegman is not a climate scientist. He is a statistician. The material Mashey alleges Wegman stole comes from Raymond Bradley, who has since apparently filed an official complaint with George Mason University. Is Mashey accusing Wegman of falsely representing himself as an expert in climate science? Is his intent to use as intellectual property ideas generated by Bradley for his own profit?
This plagiarism claim is very strange. In Mashey’s report, he seems to go out of his way to discourage readers from actually looking at either Bradley’s text or Wegman’s. Mashey writes,
“Skeptical readers are welcome to check all 35 pages, but I suspect most will read no more than few before the repetitive style gets tiring. I had to do this to gather and summarize the data. Most people need not.”
Actually, Mr. Mashey, if you want people to believe you, most people indeed need to.
On page 19 of Mashey’s report (PDF – warning large download) is the first example of Wegman’s ‘plagiarism.’ Wegman writes on page 69,
“Overall the network includes 112 proxies, and each series has been formatted into annual mean anomalies relative to the reference period used for this data, 1902-1980.”
When compared to MBH98, page 779, it does indeed look similar:
“The long instrumental records have been formed into annual mean anomalies relative to the 1902–80 reference period, …”
Not identical, but similar. But wait a minute. What is the context for this? This is the second paragraph of a ‘Summary of Global-scale temperature Patterns and Climate Forcing Over the Past Six Centuries’ by Michael Mann, Raymond Bradley and Malcolm Hughes (1998).
The authors are credited. The intent is to summarize what the authors wrote. The text is not identical, but to be an accurate summary would have to be similar. To say that Wegman plagiarised Bradley when he has Bradley’s name in the chapter title and is trying to summarize what Bradley said… well, I can see where this is going.
All of a sudden this 250 page, convoluted and poorly written report looks like a mountain to climb. It smells like a time-wasting replica of poorly written and un-thought out conspiracy theories.
I will be looking at this report more closely, but I will leave you with some quotes that I think show this to be the type of conspiracy theory nonsense that will have you looking for black helicopters or assorted nonsense. Here’s Mr. Mashey in full swing:
“During 2005-2006, Said was employed by Johns Hopkins University and that affiliation is the one listed on the WR. Did she do the WR work ―on her own time or was she in effect taking time from JHU teaching or research to work on the WR? If so, was this acceptable?”
(Let’s ask Gavin Schmidt, NASA employee and full time blogger at Real Climate…)
“The Federal government pays for many things. It is not obvious why {NIAA, ARL, ARO, NSWRC} seem to be paying statisticians and statistical physicists to attack climate science.”
(Maybe Mr. Mashey should take a good look again at what science is and how it works…)
“Other leadup to the WR is covered in [MAS2010], but it is worth knowing that Wegman, Said, Spencer, McIntyre, Singer, Kueter all attended a climate workshop November 14-16, 2005: www.climatescience.gov/workshop2005/participants.htm. Although I do not know if they met, it certainly seems likely.”
(And what did they have affixed in their lapel buttons?)
In the past few months we have seen a number of amateurish attempts to counter skeptical arguments that gained traction in what public space there is for matters climatic and anti-climactic. Today we get introduced to John Mashey’s attempt to smear Edward Wegman and reclaim the Hockey Stick for further usage. It’s definitely anti-climactic.
Climategate had a huge impact on public opinion regarding the probity of some of the scientists involved. The leaked emails clearly showed bad and bullying behaviour that left a stench in any honest reader’s nostrils. Anthony Montford, among others, chronicled Climategate and the events leading up to it in a clear, detailed narrative called The Hockey Stick Illusion. It has been praised by reviewers, including climate scientist Judith Curry. I have read it. It is good. It is accurate. I recommend it without reservation.
So, a few months ago a website called Scholars and Rogues published an incredibly lame attempt by Brian Angliss to show why nobody needed to read The Hockey Stick Illusion, citing the low number of emails that were leaked as evidence that we didn’t have enough evidence. When Steve Mosher pointed out that a crooked accountant probably had numerous accurate transactions to his credit, but that only one was needed to prove him criminal, Mr. Angliss and Scholars and Rogues sort of went away.
The late Stephen Schneider and an IT administrator named James Prall published a paper in the Proceedings of the National Academy of Sciences claiming to prove that scientists on their side of the fence were more credible than skeptical scientists, because they had more publications. Sadly, they only searched in English, they got the names, job titles and specializations of the scientists wrong, they used incorrect analysis techniques (as explained quite brilliantly by RomanM on a post at Real Climate), and used Google Scholar, a commercial database with no published quality control measures, as opposed to any one of several available academic databases. They didn’t get Stephen Schneider’s publications accurate. But that’s okay, they have the names and pictures of those they labeled (incorrectly, in many cases) as skeptics on Prall’s website.
Now comes John Mashey, intent on the destruction of Edward Wegman’s criticism of Michael Mann and Raymond Bradley’s carefully concocted Hockey Stick Chart. Wegman was asked by a congressional committee to investigate their work. His report, fully supported by the National Academy of Sciences, was devastating, citing quite correctly the fact that random noise fed into Mann’s analysis scheme could produce a hockey stick, that they used incorrect analysis of principal components in their study, and that the community of scientists involved was so closely interlinked as to be best described as incestuous, making claims of independent verification a mockery.
Before I go any further, I should note that Mashey makes another accusation that hasn’t been picked up by the media: He accuses Steve McIntyre and Ross McKitrick of being ‘recruited, coached and promoted’ by the George Marshall Institute.
I don’t think that’s true. Mr. Mashey does not provide any documentation for this accusation and I’d like to see his evidence. I certainly hope it’s of better quality than the rest of what turns out to be drivel.
John Mashey’s first bone to pick with the Wegman report is that Congressman Barton’s staff provided source material to Wegman. It is the first item in Mashey’s ‘report,’ a 250 page diatribe.
But it is entirely normal that Wegman would ask for and Barton’s staff would provide, any relevant material to speed up the investigation. I have done investigative work for several government bodies and it is in the interests of saving the public’s time and money that papers are provided. I literally cannot understand why Mashey would make this his first point.
Also on the first page of Mashey’s report is the ‘accusation’ that one of Wegman’s associates in the investigation was a post doctoral student with one year of experience, Yasmin Said. Perhaps Mr. Mashey should take a quick look at how much experience Michael Mann had when he created the Hockey Stick that became the iconic representation of climate change to the world…
John Mashey says that Wegman plagiarised material in his report to Congress.
This is odd. Wegman is not a climate scientist. He is a statistician. The material Mashey alleges Wegman stole comes from Raymond Bradley, who has since apparently filed an official complaint with George Mason University. Is Mashey accusing Wegman of falsely representing himself as an expert in climate science? Is his intent to use as intellectual property ideas generated by Bradley for his own profit?
This plagiarism claim is very strange. In Mashey’s report, he seems to go out of his way to discourage readers from actually looking at either Bradley’s text or Wegman’s. Mashey writes, “Skeptical readers are welcome to check all 35 pages, but I suspect
most will read no more than few before the repetitive style gets tiring. I had to do this to gather and summarize the data. Most people need not.”
Actually, Mr. Mashey, if you want people to believe you, most people indeed need to.
On page 19 of Mashey’s report is the first example of Wegman’s ‘plagiarism.’ Wegman writes on page 69, “Overall the network includes 112 proxies, and each series has been formatted into annual mean anomalies relative to the reference period used for this data, 1902-1980.” When compared to MBH98, page 779, it does indeed look similar: “The long instrumental records have been formed into annual mean anomalies relative to the 1902–80
reference period, …”
Not identical, but similar. But wait a minute. What is the context for this? This is the second paragraph of a ‘Summary of Global-scale temperature Patterns and Climate Forcing Over the Past Six Centuries’ by Michael Mann, Raymond Bradley and Malcolm Hughes (1998).
The authors are credited. The intent is to summarize what the authors wrote. The text is not identical, but to be an accurate summary would have to be similar. To say that Wegman plagiarised Bradley when he has Bradley’s name in the chapter title and is trying to summarize what Bradley said… well, I can see where this is going.
All of a sudden this 250 page, convoluted and poorly written report looks like a mountain to climb. It smells like a time-wasting replica of poorly written and un-thought out conspiracy theories.
I will be looking at this report more closely, but I will leave you with some quotes that I think show this to be the type of conspiracy theory nonsense that will have you looking for black helicopters or assorted nonsense. Here’s Mr. Mashey in full swing:
“During 2005-2006, Said was employed by Johns Hopkins University and that affiliation is the one listed on the WR. Did she do the WR work ―on her own time or was she in effect taking time from JHU teaching or research to work on the WR? If so, was this acceptable?”
(Let’s ask Gavin Schmidt, NASA employee and full time blogger at Real Climate…)
“The Federal government pays for many things. It is not obvious why {NIAA, ARL, ARO, NSWRC} seem to be paying statisticians and statistical physicists to attack climate science.”
(Maybe Mr. Mashey should take a good look again at what science is and how it works…)
“Other leadup to the WR is covered in [MAS2010], but it is worth knowing that Wegman, Said, Spencer, McIntyre, Singer, Kueter all attended a climate workshop November 14-16, 2005: www.climatescience.gov/workshop2005/participants.htm. Although I do not know if they met, it certainly seems likely.”
(And what did they have affixed in their lapel buttons?)
Steven Mosher,
The link references two white papers produced by MM for the GMI and slides for a talk which appears to have been given at the behest on the GMI. I would say that is enough.
You can get almost any thesis from University Microfilms for about $75. If you live in Washington, you can look at the microfilm at the Library of Congress. Enjoy
Howard U. was rated by Newsweak as the country’s second worst college.
Is this Howard University’s “Eli Rabett“?
bigcitylib
October 10, 2010 3:56 pm
Anthony, they mentioned me on air. But it isn’t like they contacted or interviewed me. I was just a pawn in their Bigfoot extravaganza.
More generally, as you know I write alot about the pathologies of the Conservative mindset. At some very deep level, these almost always involve Bigfoot, and Roswell. Hence I write about Bigfoot and Roswell. Bigfoot is the guy flying the black helicopter that ferries fat Al Gore to his meetings with the Homosexual Agenda at Roswell. Ask Jeff Id.
Cheers,
DCC
October 10, 2010 4:32 pm
Brian Angliss said:
BTW [Tom], you never addressed your fundamental inconsistency regarding the CRU emails, namely claiming in one place that the emails didn’t change the science, yet claiming that some of the emails cast the science in doubt. You can’t have it both ways, Tom.
I am flabbergasted by how poorly the AGW proponents here reason. First it’s an absurd 250 pages of drivel charging plagiarism where plagiarism is irrelevant and now somebody apparently cannot understand plain English – or write a short, coherent posting. There is absolutely no inconsistency in those two statements. The Climategate emails did not in and of themselves change anything in the so-called science. But they certainly cast doubt about the quality of the self-styled climate scientists and therefore cast doubt on the quality of their product, the science itself, just as the nonsensical arguments in this thread are doing. Is there no one with common sense among the supporters of AGW? Well, maybe Judith Curry and a handful of others. Too bad they aren’t the vocal ones.
Smokey says:
October 10, 2010 at 3:29 pm
Is this Howard University’s “Eli Rabett“?
Nice one Smokey! You just gotta love the internet!
Well then Bugs. Any comment?
Hilary Ostrov (aka hro001)
October 10, 2010 6:13 pm
DCC said:
“I am flabbergasted by how poorly the AGW proponents here reason.”
I believe it’s better known as “post-normal” reasoning 😉
Richard Sharpe
October 10, 2010 6:28 pm
Smokey says on October 10, 2010 at 3:29 pm
Howard U. was rated by Newsweak as the country’s second worst college.
Is this Howard University’s “Eli Rabett“?
Smokey, Josh Halpern might be the worlds biggest wanker for all I know, but it is a poor carpenter who blames his tools and a poor student who blames his teacher.
Well usually Eli ignores Smokey, but for the record, Howard rates 104 on US New’s list of ~300 national universities, somewhat above George Mason@ur momisugly 143. Howard is tied with Florida State, Samford (not Stanford, dear lord), the University of Kansas, Nebraska Lincoln, University of New Hampshire and the University of Tennessee. Our football team sucks FWIW.
The Rabett has no idea what strange list Smokey is quoting from, but that’s the way it is.
Bernard J.
October 10, 2010 8:09 pm
Mr Tom Fuller.
I think you will find that it’s Dr Mashey.
Bernard J, glad to see you got a guest pass out of Deltoid–enjoy the outside world. It works a little differently out here, so make sure you find an adult to escort you around.
As for Mr. Mashey, doctor is as doctor does.
Lazar
October 11, 2010 4:15 am
“was devastating, citing quite correctly the fact that random noise fed into Mann’s analysis scheme could produce a hockey stick”
No, Tom, that’s not “devastating”. Go and read Huybers’ comment on McIntyre & McKitrick 2005. See John Mashey’s comment…
“On p.58, under Cressie, I show the tale of his advice, including a link to the file he sent me, and a 1-line paraphrase:
“I concur with the MBH decentering issue, so compute it the _right_ way”
What he meant was, in effect, do the same as Wahl&Amman did,. How do I know what he meant? Because I asked him.
Anybody who knows anything about this sort of statistics knows perfectly well that there may be an error in a procedure, but whether or not it makes any appreciable difference depends on the data and parameters.”
Lazar, that is lame:
“Anybody who knows anything about this sort of statistics knows perfectly well that there may be an error in a procedure…”
Lay out the code, and show us that specific “error in procedure.”
And regarding your link, the comments following Mashey’s debunk him.
Lazar
October 11, 2010 5:07 am
“show us”
no
Bernard J.
October 11, 2010 5:09 am
It works a little differently out here,
Which is why I usually stick to discussions that are based upon at least a minimum standard of science. Nevertheless, I occasionally venture into swamps against my better judgement, in the hope that I might make a simple point or two. It’s up to the porcines to recognise the nacre.
…so make sure you find an adult to escort you around.
Gee, I’ll have to go back to Deltoid then and locate one who might assist.
But seriously, for just a moment Fuller, it serves no-one any good to use a gutter press approach to colouring a piece to which one attributes one’s own name. What’s the reason for your reluctance to acknowledge Mashey’s considerable competence and education in analysical process?
Tabloid is as tabloid does, I guess.
[Archived for posterity]
Lazar, let’s give the mic to Brandon Shollenberger at CaS:
#188 Alex Harvey, it is worse than you say. You can download Noel Cressie’s e-mail to Edward Wegman from a link on page 58 (or 134). It does not support John Mashey’s claims at all. The paraphrase he gives on page 58, “I concur with the MBH decentering issue, so compute it the ―right way,” is complete rubbish.
First, Cressie specifically state he “concurs” with the technical contents of the Wegman Report. Second, when we look at Cressie’s second point (page 134 tells us the number), we see he actually agrees with Wegman. Wegman criticized decentered PCA, even saying technically it isn’t even PCA. He then said what the proper way of doing PCA is. Cressie, in his e-mail, recommended Wegman show the results of proper PCA alongside the results of the decentered PCA. He suggested Wegman make a figure to show the difference in results, that’s all.
Somehow, Mashey has managed to take Cressie’s e-mail to say the exact opposite of what it actually says. Cressie did not support MBH’s approach to PCA, nor did it even discuss W&A’s. However, he did “concur” with the technical details of Wegman’s report, which condemned both the MBH and W&A approaches.
But apparently we are supposed to trust Mashey’s fake claims because he asked Cressie what he meant.
And Bernard, should I post here various comments from you showing why rather than receiving lessons in courtesy from you I, like everybody else with common sense, ignore you as much as possible?
juanita
October 11, 2010 7:39 am
I’m just glad I’m not the only person who thinks of Richard Dreyfuss and Teri Garr every time I look at a plate of mashed potatoes.
Thanks for the python, it was like a vitamin shot!
I tried to make this point to Tom over at Kloor’s site, but apparently Keith felt it failed to meet his moderation standard for some reason. So it goes
Tom and Steven, the ICCER final report bore out my arguments regarding insufficient context in the emails. Not only did they use nearly the same statistical argument you reject, they answered Steven’s complaints about Kieth Briffa’s emails when Briffa produced an email that hadn’t been in the CRU archive that totally changed the context of the emails that WERE in the archive.
I’m sure that you’ve both read the report, but I analyzed it in depth and pointed out where their analysis coincided with mine here.
Lazar
October 11, 2010 8:54 am
Tom Fuller,
“Lazar, let’s give the mic to Brandon Shollenberger”
… which reiterates the trivial truth of ‘decentered PCA inflates the variance of ‘hockey-stick’ shapes’… it doesn’t address impacts of this flawed methodology on the results of MBH98… linking to it is unresponsive to my comment… it doesn’t support your claim of the Wegman report is “devastating” to MBH98
like I said… read Huybers
Brian Angliss, just look at how much you x’d out of your own webpost. You were wrong. You don’t need a statistically significant number of emails to verify that emails concerned with wrongdoing exist and are damning.
Lazar, well, Wegman was devastating to MBH98 and it still is. What is flawed about the methodology of Wegman?
Lazar
October 11, 2010 1:52 pm
Tom, “Wegman was devastating to MBH98 and it still is.”
You can repeat it however many times… doesn’t make it true. “What is flawed about the methodology of Wegman?”
The flaw under discussion is in MBH98, the tendency of decentered PCA to inflate the variance of hockey stick shapes. You’re claiming that fact is of itself “devastating” to MBH98. I’m claiming that’s flawed logic, as Mashey wrote… “there may be an error in a procedure, but whether or not it makes any appreciable difference depends on the data and parameters”… in other words the seriousness of the flaw depends on how it effects the results. A flaw may or may not have a big effect on results, but the mere presence of a flaw is not “devastating”.
Read Huybers,
Actually, Tom, the post you linked to is not the source of my statistical argument, this post is. You linked to a post where I attempted to document flaws in logic by McIntyre, yourself, Mosher, and many others. And it was only some of the criticisms of McIntyre that I struck out. I didn’t have to do that with the arguments I made relating to you, Mosher, or even all of my criticisms of McIntyre. I admit that I made some errors – that’s why I struck those out. But my statistical argument was not one of the things that required correction. Not by a long shot.
As I said then, it’s not that you need a statistical sample so much as it is that you need a sample that hasn’t been self-selected by some unknown party according to some unknown (and likely automated) criteria. We all knowingly laugh at web polls at sites like CNN because they are so easily freeped that they can never be considered representative of anything other than the people who decided to vote.
We should similarly laugh at the idea that the emails represented anything resembling the full context of the emails. And the ICCER final report provided at least a half-dozen examples of how the full context was not available. Allow me to refer you to Section 8.5 of the ICCER report for the best example thereof, where Briffa produced a set of emails that proved he did nothing wrong while he was working as editor of Holocene.
JFA in Montreal
October 11, 2010 4:23 pm
LOL ! “Pay no attention to that man behind the curtain.” 🙂
Another Ian
October 12, 2010 6:55 pm
Re
curly says:
October 10, 2010 at 1:34 pm
From http://en.wikipedia.org/wiki/MIPS_Technologies
“Some notable people who worked in MIPS:David Hitz, Joseph DiNucci, James Billmaier, Steve Blank, Dan Levin, Bob Miller, Skip Stritter, John L. Hennessy, John Mashey, John P. McCaskey, Stratton Sclavos. Board members included: Bill Davidow.”
Steven Mosher,
The link references two white papers produced by MM for the GMI and slides for a talk which appears to have been given at the behest on the GMI. I would say that is enough.
You can get almost any thesis from University Microfilms for about $75. If you live in Washington, you can look at the microfilm at the Library of Congress. Enjoy
Howard U. was rated by Newsweak as the country’s second worst college.
Is this Howard University’s “Eli Rabett“?
Anthony, they mentioned me on air. But it isn’t like they contacted or interviewed me. I was just a pawn in their Bigfoot extravaganza.
More generally, as you know I write alot about the pathologies of the Conservative mindset. At some very deep level, these almost always involve Bigfoot, and Roswell. Hence I write about Bigfoot and Roswell. Bigfoot is the guy flying the black helicopter that ferries fat Al Gore to his meetings with the Homosexual Agenda at Roswell. Ask Jeff Id.
Cheers,
Brian Angliss said:
I am flabbergasted by how poorly the AGW proponents here reason. First it’s an absurd 250 pages of drivel charging plagiarism where plagiarism is irrelevant and now somebody apparently cannot understand plain English – or write a short, coherent posting. There is absolutely no inconsistency in those two statements. The Climategate emails did not in and of themselves change anything in the so-called science. But they certainly cast doubt about the quality of the self-styled climate scientists and therefore cast doubt on the quality of their product, the science itself, just as the nonsensical arguments in this thread are doing. Is there no one with common sense among the supporters of AGW? Well, maybe Judith Curry and a handful of others. Too bad they aren’t the vocal ones.
Smokey says:
October 10, 2010 at 3:29 pm
Is this Howard University’s “Eli Rabett“?
Nice one Smokey! You just gotta love the internet!
Well then Bugs. Any comment?
DCC said:
I believe it’s better known as “post-normal” reasoning 😉
Smokey says on October 10, 2010 at 3:29 pm
Smokey, Josh Halpern might be the worlds biggest wanker for all I know, but it is a poor carpenter who blames his tools and a poor student who blames his teacher.
Well usually Eli ignores Smokey, but for the record, Howard rates 104 on US New’s list of ~300 national universities, somewhat above George Mason@ur momisugly 143. Howard is tied with Florida State, Samford (not Stanford, dear lord), the University of Kansas, Nebraska Lincoln, University of New Hampshire and the University of Tennessee. Our football team sucks FWIW.
The Rabett has no idea what strange list Smokey is quoting from, but that’s the way it is.
Mr Tom Fuller.
I think you will find that it’s Dr Mashey.
Bernard J, glad to see you got a guest pass out of Deltoid–enjoy the outside world. It works a little differently out here, so make sure you find an adult to escort you around.
As for Mr. Mashey, doctor is as doctor does.
No, Tom, that’s not “devastating”. Go and read Huybers’ comment on McIntyre & McKitrick 2005. See John Mashey’s comment…
Lazar, that is lame:
“Anybody who knows anything about this sort of statistics knows perfectly well that there may be an error in a procedure…”
Lay out the code, and show us that specific “error in procedure.”
And regarding your link, the comments following Mashey’s debunk him.
“show us”
no
Which is why I usually stick to discussions that are based upon at least a minimum standard of science. Nevertheless, I occasionally venture into swamps against my better judgement, in the hope that I might make a simple point or two. It’s up to the porcines to recognise the nacre.
Gee, I’ll have to go back to Deltoid then and locate one who might assist.
But seriously, for just a moment Fuller, it serves no-one any good to use a gutter press approach to colouring a piece to which one attributes one’s own name. What’s the reason for your reluctance to acknowledge Mashey’s considerable competence and education in analysical process?
Tabloid is as tabloid does, I guess.
[Archived for posterity]
Lazar, let’s give the mic to Brandon Shollenberger at CaS:
#188 Alex Harvey, it is worse than you say. You can download Noel Cressie’s e-mail to Edward Wegman from a link on page 58 (or 134). It does not support John Mashey’s claims at all. The paraphrase he gives on page 58, “I concur with the MBH decentering issue, so compute it the ―right way,” is complete rubbish.
First, Cressie specifically state he “concurs” with the technical contents of the Wegman Report. Second, when we look at Cressie’s second point (page 134 tells us the number), we see he actually agrees with Wegman. Wegman criticized decentered PCA, even saying technically it isn’t even PCA. He then said what the proper way of doing PCA is. Cressie, in his e-mail, recommended Wegman show the results of proper PCA alongside the results of the decentered PCA. He suggested Wegman make a figure to show the difference in results, that’s all.
Somehow, Mashey has managed to take Cressie’s e-mail to say the exact opposite of what it actually says. Cressie did not support MBH’s approach to PCA, nor did it even discuss W&A’s. However, he did “concur” with the technical details of Wegman’s report, which condemned both the MBH and W&A approaches.
But apparently we are supposed to trust Mashey’s fake claims because he asked Cressie what he meant.
And Bernard, should I post here various comments from you showing why rather than receiving lessons in courtesy from you I, like everybody else with common sense, ignore you as much as possible?
I’m just glad I’m not the only person who thinks of Richard Dreyfuss and Teri Garr every time I look at a plate of mashed potatoes.
Thanks for the python, it was like a vitamin shot!
I tried to make this point to Tom over at Kloor’s site, but apparently Keith felt it failed to meet his moderation standard for some reason. So it goes
Tom and Steven, the ICCER final report bore out my arguments regarding insufficient context in the emails. Not only did they use nearly the same statistical argument you reject, they answered Steven’s complaints about Kieth Briffa’s emails when Briffa produced an email that hadn’t been in the CRU archive that totally changed the context of the emails that WERE in the archive.
I’m sure that you’ve both read the report, but I analyzed it in depth and pointed out where their analysis coincided with mine here.
Tom Fuller,
… which reiterates the trivial truth of ‘decentered PCA inflates the variance of ‘hockey-stick’ shapes’… it doesn’t address impacts of this flawed methodology on the results of MBH98… linking to it is unresponsive to my comment… it doesn’t support your claim of the Wegman report is “devastating” to MBH98
like I said… read Huybers
Brian Angliss, just look at how much you x’d out of your own webpost. You were wrong. You don’t need a statistically significant number of emails to verify that emails concerned with wrongdoing exist and are damning.
Lazar, well, Wegman was devastating to MBH98 and it still is. What is flawed about the methodology of Wegman?
Tom,
“Wegman was devastating to MBH98 and it still is.”
You can repeat it however many times… doesn’t make it true.
“What is flawed about the methodology of Wegman?”
The flaw under discussion is in MBH98, the tendency of decentered PCA to inflate the variance of hockey stick shapes. You’re claiming that fact is of itself “devastating” to MBH98. I’m claiming that’s flawed logic, as Mashey wrote… “there may be an error in a procedure, but whether or not it makes any appreciable difference depends on the data and parameters”… in other words the seriousness of the flaw depends on how it effects the results. A flaw may or may not have a big effect on results, but the mere presence of a flaw is not “devastating”.
Read Huybers,
Actually, Tom, the post you linked to is not the source of my statistical argument, this post is. You linked to a post where I attempted to document flaws in logic by McIntyre, yourself, Mosher, and many others. And it was only some of the criticisms of McIntyre that I struck out. I didn’t have to do that with the arguments I made relating to you, Mosher, or even all of my criticisms of McIntyre. I admit that I made some errors – that’s why I struck those out. But my statistical argument was not one of the things that required correction. Not by a long shot.
As I said then, it’s not that you need a statistical sample so much as it is that you need a sample that hasn’t been self-selected by some unknown party according to some unknown (and likely automated) criteria. We all knowingly laugh at web polls at sites like CNN because they are so easily freeped that they can never be considered representative of anything other than the people who decided to vote.
We should similarly laugh at the idea that the emails represented anything resembling the full context of the emails. And the ICCER final report provided at least a half-dozen examples of how the full context was not available. Allow me to refer you to Section 8.5 of the ICCER report for the best example thereof, where Briffa produced a set of emails that proved he did nothing wrong while he was working as editor of Holocene.
LOL ! “Pay no attention to that man behind the curtain.” 🙂
Re
curly says:
October 10, 2010 at 1:34 pm
From
http://en.wikipedia.org/wiki/MIPS_Technologies
“Some notable people who worked in MIPS:David Hitz, Joseph DiNucci, James Billmaier, Steve Blank, Dan Levin, Bob Miller, Skip Stritter, John L. Hennessy, John Mashey, John P. McCaskey, Stratton Sclavos. Board members included: Bill Davidow.”