Hump Day Hilarity – The 97% climate expertise [warning graphic content]

Josh writes:

I had an idea with Brandon [Shollenberger] about discovering a secret stupid chart along the lines of the one he found.

Shollenberger writes:


I think I may have found the strangest chart I have ever seen. You can see it below, taken from the newly published paper on the supposed “consensus on the consensus” on global warming:

consvexpertise2

Now, I discussed this paper a bit yesterday, and there are probably a lot of things more important to discuss than this chart. Those other things aren’t as funny though. You see, this chart is complete nonsense. Look at the x-axis. See how it says “Expertise”? Tell me, what scale do you think that’s on?

You’re wrong. It doesn’t matter what your answer might have been; it’s wrong. It’s wrong because there is no scale for the x-axis on this chart.

Seriously. This is what the authors of the paper had to say about the chart:

Figure 1 uses Bayesian credible intervals to visualise the degree of confidence of each consensus estimate (largely a function of the sample size). The coloring refers to the density of the Bayesian posterior, with anything that isn’t gray representing the 99% credible interval around the estimated proportions (using a Jeffreys prior). Expertise for each consensus estimate was assigned qualitatively, using ordinal values from 1 to 5. Only consensus estimates obtained over the last 10 years are included.

See: http://www.hi-izuru.org/wp_blog/2016/04/strangest-chart-ever-created/

That essay from Shollenberger is well worth a read as it shows how ridiculous the new paper from John Cook actually is. Following Cook’s lead, and in turn, working with Josh, we were able to bin the major players on the climate and consensus wars.

97pct-expertise-chart

Here is the rationale for the binning from our “independent” raters:

Ehrlich and Stern don’t seem to have any particular direction, they throw a lot of alarmist mud at random times, hence “confused”. Stern earns a  lot more than Ehrlich so rates higher on the $10K units.

Nuccitelli and Allen seem unaffected by any facts, and drone on with the same message, just repackaged. Hence the “dim” bin.

McKibben is a nice guy, but has only minimal climate expertise, basically regurgitating news items and seeing the effects of every weather event as “proof” of global warming on his Twitter feed. He also thinks we can “change physics” so he gets binned in the “idiot” category. Mandia, who has meteorological expertise, would not be in this category at all, except that he self binned by becoming a climate superhero. Klein writes bestselling books, that she herself doesn’t quite understand, but assures us she’s “changing everything“, so fits in this category as a high-earner.

Lew and Oreskes, self binned, are permanently sealed in, with no escape from their own seepage in their self-made conspiracy theory theater. Oh, and your pets will die too.

Schmidt, has a category all his own, but often disappoints because his wish is to never debate a skeptic or answer a question from a skeptic.

Trenberth and Hansen do a lot of projection, seeing missing heat, coal death trains, and disappearing cities under sea level rise all due to climate model output.

Cook, caught lying on more than one occasion. Low earner.

Gleick, admits his own fraud, low earner, about to be even lower.

Gore, a high earner raking in millions, can’t even do high school science without faking results in post-production.

Mann, is an outlier. In keeping with McIntyre’s note about reporting both positive and negative results, we report the outlier here. Enough said.

Note: Bob Ward of the Grantham Institute was also on this list for plotting, but because at one time or another he has put himself in every bin, we had to discard him and his data as being “unbinnable”.

0 0 votes
Article Rating

Discover more from Watts Up With That?

Subscribe to get the latest posts sent to your email.

110 Comments
Inline Feedbacks
View all comments
Marcus
April 13, 2016 2:33 pm

..OMG, that is funny ! 50 STARS !

george e. smith
Reply to  Marcus
April 13, 2016 3:56 pm

I don’t see Bill Nye on the chart; how come ??
g

Marcus
Reply to  george e. smith
April 13, 2016 4:00 pm

..They didn’t want to get banned for using swear words !

Reply to  george e. smith
April 13, 2016 4:17 pm

John Holdren is missing, too. He probably gets a pretty nice Executive Branch salary, falling well above mid-$10k expertise units, and with formal training in physics ascends somewhere into the pink zone.

Severing
Reply to  george e. smith
April 13, 2016 8:04 pm

He’s out shopping for a shrubbery…Nye!

Eugene WR Gallun
Reply to  george e. smith
April 13, 2016 8:53 pm

Severing — Too funny for words — Eugene WR Gallun

Michael 2
Reply to  george e. smith
April 14, 2016 7:07 am

Bill Nye is with Al Gore on the post-production thermometer science thing.

DD More
Reply to  george e. smith
April 15, 2016 8:43 am

George, he is off with the Knights of Ni. Just has a heavy accent and is misunderstood.
Concerning Gore, should the non-numbered “Expertise” $10,000 be noted as a Log Scale?

Santa Baby
Reply to  Marcus
April 13, 2016 9:53 pm

Social or Marxist climate science expertise?

JohnWho
April 13, 2016 2:34 pm

I think you should go back and adjust the data.
I’m sure if you torture it enough you’ll find that it forms a Hockey Stick.

Niff
Reply to  JohnWho
April 13, 2016 6:09 pm

Yup, just arrange that sequence and think up an axis title…..LOL

PaulH
Reply to  JohnWho
April 13, 2016 6:13 pm

If you blink enough times, you can actually see a hockey stick!

JohnKnight
Reply to  JohnWho
April 14, 2016 2:12 pm

Yep, me thinks Josh missed a shot at a wide open net ; )

Marcus
April 13, 2016 2:35 pm

..And I will donate to help you to defend yourselves from the coming lawsuits !! LOL

urederra
April 13, 2016 2:36 pm

What is the r² value for the regression line in the second chart? 😛

Marcus
Reply to  urederra
April 13, 2016 2:39 pm

…97% of course !!

Reply to  urederra
April 13, 2016 2:44 pm

What would you like it to be? 😉

urederra
Reply to  tarran
April 13, 2016 2:54 pm

0.97 😛

Tom in Florida
April 13, 2016 2:42 pm

Reminds me of when I attended the World Ocean Expo while stationed in Okinawa back in 1978 (could have been 1976). The Russians had an exhibit pavilion and all of their people wore name tags that titled them “Expert”. Expert at what it didn’t say and I didn’t ask (most likely intelligence gathering).

Penelope
April 13, 2016 2:46 pm

No wonder I read Hump Day Hilarity as Hillary. Makes perfect sense w the Baysian posterior.
sorry

Phillip Bratby
Reply to  Penelope
April 13, 2016 11:08 pm

No need to apologise.

Tom Halla
April 13, 2016 2:52 pm

The satire is every bit a valid as a Cook study 🙂

Louis Hooffstetter
April 13, 2016 2:59 pm

Josh nails it again!

April 13, 2016 3:02 pm

A significant portion of the authors of THIS paper, which tries to make a point about the “consensus of the consensus”, are the very authors of the of the subject papers! “We were right, and the fact that we all agree about us being right, and also that we all came up with the exact same percentage (what are the odds of that?) also proves we are right.” (My head hurts!)
Isn’t there an ethical problem here?

Robert of Ottawa
Reply to  George Daddis
April 13, 2016 3:46 pm

Certainly an terminal case of circular reasoning.

Reply to  Robert of Ottawa
April 13, 2016 5:14 pm

Another reference to a circular…..er…um….activity comes to mind….but that could very well be caused by climate change making women even less capable of participating in it…..:P

Hivemind
Reply to  Robert of Ottawa
April 14, 2016 6:01 am

Circular reasoning, like the Oozlum bird.

simple-touriste
April 13, 2016 3:06 pm

For this to have any value, you need to claim raters are independent (even if they worked as a team, discussed ratings, or if there was only one rater).

Travis Casey
April 13, 2016 3:21 pm

Hilarious!

RealOldOne2
April 13, 2016 3:25 pm

When I read “the density of the Bayesian posterior” I thought it might be referring to one of these Bay mare’s posterior, https://www.google.com/search?q=bay+mare&tbm=isch&tbo=u&source=univ&sa=X&ved=0ahUKEwiMnemc0ozMAhXpzoMKHUhNA1UQsAQIHA&biw=1280&bih=900 , as some have demonstrated similarities to portions of a horses’ anatomy.

Robert of Ottawa
Reply to  RealOldOne2
April 13, 2016 3:49 pm

I was thinking of that Bayesian babe Kim Kardashian’s posterior

April 13, 2016 3:32 pm

Beautiful satire. Post it to Cook and Mann and Oreskes. An IED equivalent blowing up their warmunist nonsense.

JGarr
April 13, 2016 3:39 pm

The average of the data points looks to be about 79.5

April 13, 2016 3:41 pm

Should Aaron Mair get a mention since he mentioned the 97% for the last 5 or 6 questions Cruz asked him?
See:

(Regulars may have seen this here several weeks ago.)

Reply to  Werner Brozek
April 15, 2016 5:59 am

That video shows an OUTSTANDING exchange. Ted Cruz embarrassed, thoroughly and rightly so, the Sierra Club president. That makes the point that I am sure many, if not all, skeptics, such as I, have experienced when asking any devout believer in the religion of global warming for the basis of their belief. Fairly recently I wrote to the head of the Green Party in Canada, Elizabeth May, asking what specific scientific evidence she has that proves that people are responsible. She did reply, but with pretty much the same sort of nonsense as that emitted by Aaron Meir. Thank you Werner for having found and posted that video exchange

Marcus
April 13, 2016 3:46 pm

..More Hump Day funnies…..( edited for safety )
tikkun613 Fox News
19 minutes ago
First They Told Me It Was Global Cooling
And I Couldn’t Stop Shivering.
Then They Told Me There Was Global Warming
And I Could Not Stop Perspiring.
Now They Tell Me It’s Climate Change
And I Can’t Figure-Out What Clothes To Wear.
They Had me So Confused,
I Suffered An Anxiety Attack.
When I Went For Counseling, I Was Told My
Insurance No Longer Covered Psychiatric Issues
But If I Covered My ACA Co-Pay Of $5,000,
I Could get Two Aspirin.
When I Called, They Told Me The
Computers Were Hacked,
My Records No Longer Existed
And I Would Need To Return The Aspirin.
-I.R.Sigel

SMC
April 13, 2016 3:51 pm

If I play connect the dots, it kind’a reminds me of a bunny rabbit. 🙂

PiperPaul
Reply to  SMC
April 13, 2016 6:21 pm

Not an elephant wiggling its trunk?

Rich Carman
April 13, 2016 4:48 pm

When I was in grad school, we developed the “Law of Arbitrary Curves” which states that if you get enough data points spread randomly over a wide enough area, you can draw a curve with any shape you desire. This often tuned bad data days into productive graphs.

PiperPaul
Reply to  Rich Carman
April 13, 2016 6:24 pm

Sort of like constellations and Zodiak. You see what you want to see.

expat
Reply to  PiperPaul
April 14, 2016 9:54 am

Yup, Orion the Hunter is called the Plow by Brits. I always thought of it as a kite. It’s upside down in the southern hemisphere so no idea what they think of it.

Jer0me
Reply to  PiperPaul
April 15, 2016 8:09 am

Hemisphereist. It’s perfectly aligned diwn here. YOU have it upside-down!

ossqss
April 13, 2016 5:08 pm

How sad!
These people are directly assisting in setting global energy policy, remember the POTUS references it frequently along with a large list of other formal positions, down to your community level.
These are not ill informed people, however I suspect the discussions with their conscience, everyday, is not a good one to experience.

TDBraun
April 13, 2016 5:31 pm

The way they try to hide their “assigned qualitatively” value at the end of a jargon-ridden paragraph is deliberate obfuscation, meant to deceive.

Santa Baby
Reply to  TDBraun
April 13, 2016 10:00 pm

It’s just policy based propaganda.

April 13, 2016 5:42 pm

Brandon/Josh,
Outstanding . . . both of you.
John

brantc
April 13, 2016 5:47 pm

They had to throw “credible” in there just to make sure….

Seth
April 13, 2016 6:41 pm

Look at the x-axis. See how it says “Expertise”? Tell me, what scale do you think that’s on?

It doesn’t look like it’s on a numerical scale. It looks like the studies are in order of the expertise of the groups sampled by the study.
It’s a bar chart, but with margin of error shown rather than colouring in the bar.

You’re wrong. It doesn’t matter what your answer might have been; it’s wrong. It’s wrong because there is no scale for the x-axis on this chart.

No, I was right. Bar charts aren’t as esoteric as you think.

Reply to  Seth
April 13, 2016 6:59 pm

Seth;
It’s a bar chart, but with margin of error shown rather than colouring in the bar.
May I refer you to:
Expertise for each consensus estimate was assigned qualitatively
Can you explain to me, using small words because I’m really stupid, how one calculates a margin of error for a qualitative measurement?

Reply to  davidmhoffer
April 13, 2016 8:58 pm

The order on the x-axis is the ranking of the 16 groups by expertise (as judged). It’s comparable in intent to psychometric rankings of sweetness, sound loudness, and such. For each group, instead of a bar with frequentist intervals at the top edge, the box covers the 99% Bayesian credible interval. What’s missing, that would be present in a good modern psychometric study (or, study of reliability psychiatric diagnosis or symptomatology), is any measurement of the stability (reproducibility, inter-rater reliability) of this ranking across different sets of equally-qualified people tasked to perform the ranking.

Seth
Reply to  davidmhoffer
April 14, 2016 12:33 am

davidmhoffer wrote: Can you explain to me, using small words because I’m really stupid, how one calculates a margin of error for a qualitative measurement?

You don’t.
The margin of error is against the y-axis, which is quantitative.
The x-axis is the one on the bottom, which just has the different studies equally spaced.
As often occurs in charts:
http://www.wri.org/sites/default/files/uploads/per_capita_emissions.png
The y-axis is the one on the side. In the graph under discussion the margin of error against that axis is shown by colouring each 1% band with the colour of the probability of the result falling in that 1% band according to the colour chart on the right hand side.
Does that help?

M Courtney
Reply to  davidmhoffer
April 14, 2016 2:49 am

Seth, you’re illustration with the bar chart shows exactly why the “expertise” bar chart is fundamentally wrong.
On your illustration the bars are ordered from highest to lowest. The sequence is set by the y-axis.
On the fundamentally-flawed Cook graphic the bars are ordered qualitatively by Cook. He guesses at expertise. The sequence is set by his own biases.
Thus the peculiar chart doesn’t show anything meaningful (as it is based on Cook’s prejudices which are irrational). But also it is misleading as it shows a sequence and apparent trend that isn’t there. The order of bars is based on the arbitrary x-axis not the measured y-axis (this is the opposite of your illustration).
It is deceptive. It is created to present an apparent trend that isn’t there. And it managed to fool you.
So how many other people have been fooled by this work (and, think about it, by Cook’s other works).

Reply to  davidmhoffer
April 14, 2016 5:01 am

Does that help?
Yes it does. Since the quality rating was only 1-5, there can’t be 16 positions on the chart. Yet there are. I read the methodology in detail. It reads nothing more that a bewildering array of excuses to rationalize their ratings, including of course giving higher ratings to papers that the authors wrote themselves than to papers written by other people. No bias or self interest in that (SARC!). Then I found a table that bins all the results into the supposed bar chart you say is there, but only has results in four bins. One of them is empty (bin 4). Plus, the results in bin 5 cover over half the horizontal axis on the chart, with 4 missing and 1,2, and 3 occupying the rest.
So you answered the question with enough expertise to suggest you understand these issues, which just makes you a defender of a study that I suspect the fr__dulent term would not be sufficient to express the level of depravity, misinformation and misdirection that have been incorporated into this study which doesn’t amount to a total piece of sh*t, but to a total piece of “evil”.
Of all the disgusting pieces of utter cr*p passed off as actual science that I have ever seen in my life, this is a whole new order of magnitude of ignorant and willful deception in support of an agenda. I don’t know if there is a heaven or h*ll, but I hope there is, because you and the people who have conspired to promote this piece of trash richly deserve to spend all eternity in the latter. Just my humble opinion of course.

seth
Reply to  davidmhoffer
April 14, 2016 3:07 pm

Yes it does. Since the quality rating was only 1-5, there can’t be 16 positions on the chart.

This is why I gave you an example of a bar chart. There’s one position for each item being graphed. Yes it does. Since the quality rating was only 1-5, there can’t be 16 positions on the chart.
In the paper it is individual estimates of the consensus. There can be however many are included in the chart.

Reply to  davidmhoffer
April 14, 2016 6:22 pm

Since the quality rating was only 1-5, there can’t be 16 positions on the chart.
And yet, there are.
Have you no shame?

JonasM
April 13, 2016 7:34 pm

Schmidt, has a category all his own, but often disappoints because his wish is to never debate a skeptic or answer a question from a skeptic.

I see what you did there.

Reply to  JonasM
April 14, 2016 2:48 am

Schmidt can talk about models and adjustments, but will avoid the physical real world. I was actually amazed at how much he doesn’t know about the sciences. Some of the stuff he says is off the wall.

April 13, 2016 8:00 pm

While I am waiting for Seth to explain how to calculate a margin of error on a qualitative measurement, can someone explain the scale on the right hand side of the graph?

GTL
Reply to  davidmhoffer
April 13, 2016 8:54 pm

Since you receive more grant money for higher temperature predictions I assume it is a Celcius temperature scale.

Neil Jordan
Reply to  davidmhoffer
April 13, 2016 9:02 pm

Easy. First you have to log-transform the y-axis. For example, take the second bar from the top, “Lying”. This reference explains how to work with log-transform lying:
http://www.appliedanimalbehaviour.com/article/S0168-1591%2810%2900054-7/abstract
Next you have to mine the abstract for suitable sciency and mathy words:
“On the basis of log–survivorship plots, frequency distributions of (log-transformed) lying episode lengths and analysis of the correspondence between recorded lying episodes and video footage of lying behaviour, a minimum lying bout criterion of 4 min was indicated.”
Next, repeat for the other bars except “Puzzled”, which takes on negative values. For “Puzzled”, you have to take logarithms of negative numbers. Root through your drawer with the calculators and choose one that gives you a number. My HP41CX displays DATA ERROR. That won’t do. My HP35S displays LOG(NEG). That won’t do either. Digging deeper, I find my HP42S. For a “Puzzled” value of -3, for example, I get 0.48 i1.36, an imaginary number perfectly suited for climate science.

Neil Jordan
Reply to  Neil Jordan
April 13, 2016 9:25 pm

Oops. I forgot the most important number of all, log -97%. That comes out to -0.013228 i1.36

Reply to  Neil Jordan
April 14, 2016 1:07 pm

NJ
My bad.
I read your 2nd paragraph as
Next you have to mime the abstract for suitable sciency and mathy words – and thought ‘No Way’ – because that would give the French a monopoly . . .
Auto

LdB
Reply to  davidmhoffer
April 13, 2016 9:11 pm

Clearly it’s the level of the color “Purple” from a start point of a color “Orange” for each qualitative measured expert.

Seth
Reply to  davidmhoffer
April 14, 2016 12:41 am

“Right colour bar indicates posterior density of Bayesian 99% credible intervals.”