Thursday, November 03, 2011
The more alert students
Sort of annoying to see the Carbon Cultists and the realists constantly bickering over the statistical methods used in averaging and adjusting a figure called the Global Temperature.
Before you start arguing about the best way to form this Global Temperature, you have to ask whether it's valid to form a Global Temperature. Validity is not an assumption you can take as given; it's a hypothesis you need to prove first.
(Of course the Carbon Cult doesn't bother to ask any questions. They simply observe each instance of increase and declare with Prophetic Certainty; no evidence needed; that the increase is caused by Carbon. )
What struck me just now: Averaging temperatures at two locations to make a single temperature is a really basic statistical error that I learned about in junior high science class.
Remember this? The teacher turns on a cold faucet and a hot faucet; puts her right hand under the cold faucet and her left hand under the hot. Says "My right hand is at 40 degrees and my left hand is at 110 degrees. Therefore my average hand is at 75 degrees, which is Just Right!"
The more alert students get the point immediately and laugh.
= = = = =
The only valid thing you can do is to examine the trends of the various locations. If all trends run up and down together, then you can start asking questions about some influence that might affect all of them. And when you ask those questions, you'll find sunspots.
But that is NOT what we have on this particular planet in this particular century. What we have is unconnected trends. (You can use this useful NOAA graphing tool to examine those trends for yourself.) Some areas show a fairly steady upward trend over the century, some areas show a fairly steady downward trend. From this real situation you can't draw any conclusions about the whole globe. You can only say that different patterns are affecting different areas. Most of those patterns may be tied to sunspots, maybe some relate to soot, some are urban heat islands, some relate to ocean currents, etc, etc.
When you start with the assumption that there is an Average Global Temperature affected by some universal global influence, you're committing the sin of circular reasoning. The assumption is not automatic. It needs to be justified independently before you try to form a number that supposedly represents all changes on the globe at once. Any "conclusion" that follows from a circular start is prima facie invalid.
= = = = =
Sidenote: Even aside from this specific piece of illogic, the arithmetic mean is very rarely an appropriate way to form a typical value. For money (income, house values) the median is best. For numbers that relate to human perception (light intensity, sound intensity) a geometric average or RMS is best.
Before you start arguing about the best way to form this Global Temperature, you have to ask whether it's valid to form a Global Temperature. Validity is not an assumption you can take as given; it's a hypothesis you need to prove first.
(Of course the Carbon Cult doesn't bother to ask any questions. They simply observe each instance of increase and declare with Prophetic Certainty; no evidence needed; that the increase is caused by Carbon. )
What struck me just now: Averaging temperatures at two locations to make a single temperature is a really basic statistical error that I learned about in junior high science class.
Remember this? The teacher turns on a cold faucet and a hot faucet; puts her right hand under the cold faucet and her left hand under the hot. Says "My right hand is at 40 degrees and my left hand is at 110 degrees. Therefore my average hand is at 75 degrees, which is Just Right!"
The more alert students get the point immediately and laugh.
= = = = =
The only valid thing you can do is to examine the trends of the various locations. If all trends run up and down together, then you can start asking questions about some influence that might affect all of them. And when you ask those questions, you'll find sunspots.
But that is NOT what we have on this particular planet in this particular century. What we have is unconnected trends. (You can use this useful NOAA graphing tool to examine those trends for yourself.) Some areas show a fairly steady upward trend over the century, some areas show a fairly steady downward trend. From this real situation you can't draw any conclusions about the whole globe. You can only say that different patterns are affecting different areas. Most of those patterns may be tied to sunspots, maybe some relate to soot, some are urban heat islands, some relate to ocean currents, etc, etc.
When you start with the assumption that there is an Average Global Temperature affected by some universal global influence, you're committing the sin of circular reasoning. The assumption is not automatic. It needs to be justified independently before you try to form a number that supposedly represents all changes on the globe at once. Any "conclusion" that follows from a circular start is prima facie invalid.
= = = = =
Sidenote: Even aside from this specific piece of illogic, the arithmetic mean is very rarely an appropriate way to form a typical value. For money (income, house values) the median is best. For numbers that relate to human perception (light intensity, sound intensity) a geometric average or RMS is best.
Labels: Blinded by Stats, Carbon Cult
¶ 5:29 PM
