Beam me to Planet Gore !
Posted on 02/21/2010 5:04:14 AM PST by mattstat
According to the stunning New York Times headline, which quoted climatologist Phil Jones, there has been no “statistically significant” global warming in the past 15 years.
Just kidding! The Times forgot to write about that. No doubt they were distracted by that golfer-guy’s TV event. Priorities!
Anyway, that’s what Mr Jones has said. Reader Francisco González has asked what that “statistically significant” means. It is an excellent question.
Answer: not much.
Here is what it absolutely, certainly does not mean: “There is a 95% chance that no warming occurred over the past 15 years.” It also does not mean: “There is a 100% chance that no warming occurred over the past 15 years.”
It also, most emphatically—slow down and read this thrice—in no way means: “We don’t know if any warming occurred.” I’ll tell you what it does mean in a minute.
It is time, now, right this minute, for the horrid term statistical significance to die, die, die (old-timers from Usenet days will grok that joke—sorry, couldn’t help myself with the second one). Nobody ever remembers what it means, and, with rare exceptions, almost everybody who uses it gets it wrong.
Statisticians have labored for nearly a century to teach the philosophy behind this term, and we just can’t make it stick. Partly it’s because the philosophy itself is so screwy; but never mind that. We must admit failure.
Here’s what “statistical significance” means in terms of global warming...
(Excerpt) Read more at wmbriggs.com ...
The issue is “clinical significance”.
There is NO GLOBAL WARMING on Earth.
For ‘statistical significance’, it would require
the data of hundreds if not thousands of Earths.
All jones was admitting was that he could not measure any warming outside of the margin of error. What he didn't say is that there is, in fact, a statistically significant cooling since 1998.
The "truth" has nothing to do with it.
Clear thinking readers should keep two things in mind.
1. You can’t “prove” anything with statistics”.
2. Figures don’t lie, but liars figure.
If something is signigficant a the 95% level that means that in a large same 95% of the events will happen. It also means that you will be dead WRONG 5% of the time. This is great if you are in the 95% of the events that you want to happen. It stinks if you are in the 5% that you don’t want to happen. Think. This treatment works on 95% of the people.
This is the biggest problem with their theories. They can’t study any earths to compare the results, so they used computer models. The models say that if they change the carbon dioxide, the temperature will rise in the future. Trouble is, even if the theory is right, it hasn’t happened yet. They claim that the prediction proves the theory.
The scientific method requires experiments that are performed many times. The results are used to come up with a formula. In this case, you’d need enough earths to ensure that the number of trials cancels out other effects that you couldn’t control. You’d increase the carbon dioxide on some of the Earths, and leave the others alone. The results would tell you what effect carbon dioxide has.
I think the author has not explained his point and his point is an excellent one. The modeling science is inadequate to make statements about statistical significance.
In order to make a claim of statistical significance, you have to have a mathematical formula that incorporates all the other variables and that this formula must work all the time. Once you know what the outcomes are supposed to be you can then determine if a change in a variable creates a change in the outcome.
In order for Jones to make a statement that there is no statistically significant climate change he believes he has a perfect understanding of climate forecasts. He obviously does not.
A note from Richard Lindzen on statistically significant warming
Look at the attached. There has been no warming since 1997 and no statistically significant warming since 1995. Why bother with the arguments about an El Nino anomaly in 1998? (Incidentally, the red fuzz represents the error ‘bars’.)
Don't blame Phil. Great site with very lively debate. What science really should be. Just read it myself. Not enough time to post.
The author knows that you cannot statistically prove a negative, then procedes to try to distract his reader from the obvious fact. A researcher uses statistics as a tool to prove his theory is true, He does not use it as a fortress demanding that he be proved wrong with a tool that is not designed to do so.
What matters is that the AGW “scientists” are trying to change the LAW, and cannot find any scientific justification for their efforts. Without such justification, their efforts amount to an attempt to wantonly destroy, block, cut-off, deny and ban the ENTIRETY of the technology of Western Civilization, thereby endangering the lives of billions of people.
So I’d say, without scientific evidence for their efforts, they are guilty of treason, attempted mass murder, and crimes against humaity, and should be executed.
- the CO2 contribution to the atmosphere from combustion is within the statistical noise of the major sea and vegetation exchanges, so a priori, it cannot be expected to be statistically significant;
The author ignores questions about the validity of the data and the method of choosing it. And he certainly knows better.
He’s also being cute about statistical significance. Statistical significance (at the 95% confidence level, or any confidence level) is a most useful concept, if judiciously applied. Statistical significance gives us a hint at the likelihood that something might be up, but it can never give us certainty.
Examples from the physical sciences abound. Millikan’s efforts to find the charge on the electron were premised on the assumption that there was one unique, true value of electron charge. The fact that repeated measurements gave differing values, did not disprove this assumption. Rather they tended to prove it because the ensemble of measurements where grouped around a central value (or some multiple of the central value), and while any one experiment might be unpredictable, if the experiment was performed a number of times, the ensemble average would always converge to the same value in a manner that was consistent with the distributions. In the language of statistics, we reject the hypothesis that the means are different.
By the late nineteenth century, astronomers had been tracking the orbit of planets, Mercury in particular, long enough and with enough accuracy to say with certainty that his orbit was inconsistent with Newton’s law of gravity and the observed distribution of mass in solar system. This was certain beyond doubt, because astronomers had mastered (some say invented) the theory of probability and knew that the anomaly, something in the order of 40 seconds of arc per century in precession of the perihelion, could not be explained by errors in the observations nor the known laws of physics.
Something else must be up.
But what was it? Astronomers were certain that all would be explained by the discovery of another planet, the hypothetical vulcan, orbit the sun inside the orbit of Mercury. Yet, the most exquisite and sensitive observations failed to find it. Einstein said that the most gratifying moment in his life was when he applied his new theory of General Relativity to orbit of Mercury and found that it accounted precisely for the observed precession of the perhelion.
Without a sound theory of probability and a knowledge of the propagation of errors, astronomers could have attributed the observed anomalies to observational errors or just assumed that “stuff happens”.
Probability theory is valuable because it tells us how certain we should be about our models and theories. The author is correct: Climate theories will be useful when they make falsifiable but true predictions about future events. Unfortunately, for AGW theorists, most of their falsifiable predictions have been just that, falsified. When their predictions fail, they confidently patch their theory and go on. And on, and on and on.
They actually have a very difficult task summarized in this extract below.
Climate studies cannot use formal statistical experimental design. For example, it is not possible to randomize the sequence in which conditions (such as atmospheric carbon dioxide concentrations) arise. Furthermore, successive observations are correlated. For example, both the temperature this year, and the carbon dioxide concentration this year depend, in part, on the temperature and concentration last year. In laboratory practice, such experiments would be randomized in some way. Failure to randomize the sequence of experiments is well known to introduce spurious correlations, or to conceal real effects. The problem with climate change (and release of carbon dioxide) is that we have a large, uncontrolled experiment. Living in a large uncontrolled experiment is potentially hazardous. (It would be unlikely to be approved by a medical ethics committee). From the statistical analysis viewpoint, we can deceive ourselves at the amount of data that we have. We have millions of day-by-day and hour-by-hour weather measurements. Ice cores give us thousands of climate measurements, extending over hundreds of thousands of years. However, it is uncertain how many independent measurements these data furnish. (We cannot simply count the number of data points to provide us with the statistical degrees of freedom). Climate measurements are correlated sequentially in time and spatially in distance. Thus, it is impossible to undertake formal statistical studies on the correlation between variables. We can undertake the correlations, but we are uncertain as to their statistical significance.
Neither can astronomy. When politics drives astromony you get the inquistition. When politics drives climate research you get Global Warming, or some other grant seeking pending catastrophe.
http://www.youtube.com/watch?v=FOLkze-9GcI
Watch these videos. Best explanation debunking the modeling.
Says you, Lonesome!
Tell you what. Find just one thing wrong with the article and prove it. (It might be easier posting your finding on the original site; so that it will be found).
And to the gent above who said “statistical significance” means 95% of large sample statements will be true. Not so. That percentage is only reached in the limit, not “large” samples, which can never be large enough to approximate that limit.
Plus, they are all, as the original article contends, conditional on the model being true. Is it? I mean, is Phil Jones model true?
His model seems to be - demonstrate that the Earth is warming significantly with noisy land temperature readings. Actually, an impossible task because it can only be proved with statistically significant reliability with heat content readings, not temperature data. And his model is false, according to satellite data and ocean buoy data. But he ignores those. In the real scientific world, those are called controls and they cannot be ignored. Obfuscating these facts with statistical methods that cannot even be applied reliably, wont change reality. Face it, your hero is a dupe and he knows it.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.