Buy
Books
Click images for more details
Currently discussing
-
Nov 19 - Barbie Jacket
on
Need an economic adviser? Here are 5 ideas for finding the very best one -
Nov 11 - US Design Hub
on
Launched my very own online store.
Links
A few sites I've stumbled across recently....
-
Science blogs
-
Political blogs
-
Other blogs
Subscribe
Josh
Copyright © 2006-2013, Bishop Hill All rights reserved.
Reader Comments (25)
Perhaps that was what she was smoking when she came up with here theory!
Perhaps her broomstick went beyond the credibility barrier?
Ah, that's better.
Well it all reminds me of the Monty Python sketch about the woman with a new theory! http://www.youtube.com/watch?v=cAYDiPizDIs
Except, of course, that that theory was incontrovertible and well-backed with evidence!
oh come now Josh, why don't you appreciate the subtleties of hockey team science??
Clearly you are part of this "guerrilla war" on climate science, here in the "untamed jungles" of the online world.....
p.s. love the cartoon!
I think you should have included an "A" tattooed on her forehead ... for "Arrogant", or perhaps "AA" for "Arrogant Activist".
(Remember her comment: "We will not be entertaining any further correspondence on the matter.")
I well remember an instruction which had been hand written in the front of some of the school text books.
"In case of fire, throw this in"
At the time, I thought it was a joke.
Not so sure now tho.
Are her hands inspired by Homer Simpson? Appropriate, if so!
If you have 100 trees and 99 correlate with temperature, then they are quite likely good proxies and it will do no harm to the analysis to include the 1 one tree that didn’t calibrate.
But if you have 200 trees and only 99 correlate with temperature, then the 101 trees that don’t calibrate tell you that the other 99 are probably not very good proxies either.
In this case you want to include the 101 that didn’t correlate, so you don’t over estimate the confidence and under estimate the error in the result.
However if you only report the trees that do calibrate, then both groups of 99 trees will appear of equal quality when in fact the first group is very likely a much better proxy than the second.
This leads to spurious conclusions because you will then weight each group of trees equal in reliability when in fact they are not at all equal.
The problem is that climate science (hockey stick) hides the trees that don’t correlate which makes all trees look like equally good proxies, even when a high number of trees in a sample are telling you they are not reliable.
Elegantly put Fred.
I found it bizarre on CA earlier when Nick Stokes, who seems an honest bloke and apparently has some statistical expertise, couldn't take this simple point on board.
Is there any significance in the fact that Ms Gergis is quadridactyle? (I'd hate it to be thought that we sceptics can't count).
In the NIWA link I posted on the other thread, they state that tree rings can be used to indicate past climatic events including drought.
The tree proxies, mainly from NZ, were taken from areas that have widely varying climatic conditions. The west coast site of Hokitika, for example, is in temperate rainforest.
How much weight can we place on their ability to match previous temperature fluctuations then?
geoffchambers
It's Josh's trademark
The magnanimity of Steve McIntyre's current post is remarkable
http://climateaudit.org/2012/06/10/more-on-screening-in-gergis-et-al-2012/
AFAIK Mickey Mouse has always been quadridactyle too. I guess "Pinky" just doesn't come out nice in comics.
Foxgoose - I would like to know if Nick has any formal statistical training and to what level. He is clearly a bright guy but I wonder if he has actually trained in stats rather then pick up bits and pieces when he needs them. If it is the latter this can lead to major blind spots due to lack of knowledge of the foundations. IMO.
Omnologos
You’re quite right about Mickey Mouse. And Goofy and Olive Oyl too. And to think I’ve been teaching art history for 15 years and I’d never noticed. Che vergogna. But what’s the connection between Gergis and Olive? Does Karoly smoke a pipe and eat spinach?
Josh,
You need to issue a corrigendum to your cartoon because you show the shaft of the hockey stick rising toward the right!!!
John
Geoff - post Lewinsky it's difficult to make comments that can't be suggestive of the unmentionable, wrt old and un-PC stuff like the Popeye cartoons...
It's not a watercolour, so it's done via big oil.
Tsk tsk tsk!
Gergis correlates well with Olive Oyl but even humourless skeptics know that correlation is not causation.
From manicbeancounter website:
."0.09°C (±0.19°C)" I'd have thought that such a value, with such an error range, was worthy of ridicule in itself.
Quadridactyle, indeed. Evidence of cartoon evolution, emerging from Waltus Disneae stock, as has been noted.
V useful too when you dont have a lot of space, and it somehow looks right to me ;-)
Martin A
"0.09°C (±0.19°C)" I'd have thought that such a value, with such an error range, was worthy of ridicule in itself.
Jun 12, 2012 at 8:16 AM | Martin A
Martin A, good point, and as I discussed on the "Gergis Paper Disappears" thread, at the end of the online May 17, 2012 media event David Karoly went out of this way to reiterate three times (as though this is what the media should most understand) that this study establishes that there was no MWP for Australia and the past 50 years are truly unprecedented (the study is about "Australasia" but he was saying "Australia" or "Australian region" at the end of that online press conference):
=========================================================================
David Karoly gets rids of the Medieval Warm Period for Australia
Karoly: their study "shows for the first time that there was no substantial MWP in the Australian region."
Karoly: "so there wasn't a MWP affecting Australia"
Karoly: "and there has been no period in at least the last thousand years which as been as warm as the past fifty years, with a very high probability"
Smoking a rather large spliff