| « | Remember Where and When |
»
|
|
Tuesday, November 24, 2009
PDOOMA!
The government is extremely fond of amassing great quantities of statistics. These are raised to the nth degree, the cube roots are extracted, and the results are arranged into elaborate and impressive displays. What must be kept ever in mind, however, is that in every case, the figures are first put down by a village watchman, and he puts down anything he damn well pleases. -- Sir Josiah Stamp
If it cannot be expressed in figures, it is not science; it is opinion. -- Robert A. Heinlein
43.7% of all statistics are made up on the spot. – Author unknown
Among the eternal verities of the engineering trade is this:
The size of the error bar -- in layman’s terms, the amount by which the figure cited could be high or low -- determines whether the figure is worthy of consideration, as evidence or as support for some proposed solution to a problem.
- A small error bar, which indicates high confidence in the figure, enhances the authority of the figure.
- A large error bar, which indicates low confidence in the figure, renders the figure suspect and unreliable.
- An unknown or undisclosed error bar indicates that the figure is a PDOOMA.
PDOOMA, one of engineering’s most beloved acronyms, stands for “Pulled Directly Out Of My Ass.” The figure is almost certainly guesswork or worse -- in extreme cases, pure fiction. The best giveaway to a PDOOMA figure is the absence of an error bar.
The PDOOMA figure is becoming a major player in political rhetoric. Conservatives have had to cope with quite a few of them in recent years. Because we prefer not to fight too many duels -- ammunition has gotten quite expensive, y’know -- we usually refrain from calling the PDOOMA's proponent a liar. The usual conservative response to a PDOOMA from a left-liberal is simply to dispute the figure's accuracy, or perhaps its relevance. But the unprecedented proliferation of political PDOOMAs has made that approach counterproductive. As Ann Coulter said in her book Slander: Liberal Lies About The American Right, we find ourselves passing over quite a lot of leftist fiction so we can concentrate on the lies of earth-shaking magnitude.
Here are a few PDOOMAs of recent vintage:
- Al Gore's claim that, if nothing is done about "global warming," Earth will experience a sea level rise of twenty feet.
- Marian Wright Edelman's claim that every extra dollar devoted to governmentally funded day care would ultimately save $3.38 in federal and state expenditures.
- Congressional Democrats claim that $500 billion can be squeezed out of Medicare expenditures over the next ten years by "eliminating waste, fraud, and abuse."
- Barack Hussein Obama's "47 million Americans can't get health insurance."
- Barack Hussein Obama's "650,000 jobs created or saved" by the $787 billion stimulus bill.
...and not an error bar in the lot.
The most annoying aspect of these PDOOMAs is how hard it is to get their promulgators to stand their ground and defend it. When challenged, they immediately change the subject, or attack the objector -- and subsequently act as if the PDOOMA went unchallenged, and should be taken as authoritative and trustworthy. But left-liberalism being inherently fraudulent and treacherous, perhaps this is only to be expected.
The proliferation of PDOOMAs on the subject of health care has slowed down somewhat in recent weeks, but there remain a couple that animate the Democrats' drive for the nationalization of American medicine. One is the "47 million uninsured" canard. No responsible analyst would propound such a figure without also stating how he'd arrived at it, what the error bar is, and whether factors other than the politically potent implication of unaffordability play a part in a significant fraction of the total. The last of those is a particularly important question at a time when among America's most important social ills is our problem with illegal immigration.
But no field beats "global warming / climate change" for reliance on the PDOOMA. The whole "discipline" is PDOOMAs from first to last. Global temperature assessment is inherently fraught with major uncertainties. Measurements of global temperature have error bars of several Centigrade degrees. The recent revelations of wholesale distortion and data-dredging at the University of East Anglia Climatic Research Unit, apparently carried on shamelessly by hundreds, if not thousands, of the best known "global warming" flacksters in the world, should be a lesson to us about the probity of "scientific measurements" advanced to serve a political cause. Self-nominated "scientists" are just as capable of deception, including self-deception, as any other human being...especially when vast quantities of money are at stake.
If there's an irony over this originating from the Right, it would be that, since left-liberals are generally aware that their own "statistics" and "projections" are PDOOMAs, they instinctively reject figures that countervail their contentions. For example, it will do you no good in a conversation with a gun-control advocate to cite Florida's impressive decrease in certain violent crimes after it became a shall-issue state -- very nearly 80% over the first two years. He'll assume you made it up. Nor will he bother to check it for himself; it violates Holy Leftist Writ about the evil of firearms, and that's all he needs to know. Neither can you make any headway with a rabid welfarist by pointing out that the Reagan-era tax rate cuts correlated with federal revenue increases of over $390 billion per year by 1989. "Tax cut" is profanity to a statist of any stripe; you simply have to be lying.
In other words, you can't persuade a left-liberal by using facts.
Indeed, it might not be possible to persuade a left-liberal by any means available to Man. Left-liberal Causes, being quasi-religious in nature, receive devotion of religious intensity from their devotees. That's why they're unabashed about their use of PDOOMAs. The end justifies the means; no deception is out of bounds if it serves Ultimate Truth.
Depressing, especially to one who respects the connection between numbers and reality.
UPDATE: See also these maunderings over temperature and precision measurement.
Comments
Nice term, “PDOOMA”. For work requiring formal footnotes, the version I see most often is “… was obtained via rectal extraction”.
Good piece.
Posted by on 11/24/2009 at 01:47 PMGore recently said the temperature of the earth’s mantle at 2 Km down is two million degrees. Talk about a hockey stick, that’s a rise of a thousand degrees per meter. His “take away” was that geothermal engineers now have drill bits that can cope with this temperature gradient.
No, he didn’t mention an error range, he didn’t specify Celsius or Kelvin or Farenheit either, so I’m beginning to suspect this pitch is “just a little outside”.
To your point: statistics have another appeal to liberals, challenges can easily devolve into a jargon-loaded discussion of methodology, meaning the challenge is effectively kicked back to the “experts” while ordinary folk wander off in search of refreshments or light entertainment. As intended, likely.
Posted by Ol' Remus on 11/24/2009 at 03:15 PMI have always used WAG (wild-assed guess) for your PDOOMA, and SWAG (scientific WAG) for when I had at least some evidence, but not enough to really know for sure.
WAG example:
Boss: I need an estimate to develop this system, whose name alone I know and which you’ve never heard of.
Me: Seventeen years.SWAG example:
Boss: How much would ten of these servers cost?
Me: Well, I bought one six months ago for $X, so about $10X or so seems right.Posted by Jeff Medcalf on 11/24/2009 at 06:14 PMI find “brown numbers” a terser way to describe them.
Posted by Karl Gallagher on 11/24/2009 at 07:23 PMit will do you no good in a conversation with a gun-control advocate to cite Florida’s impressive decrease in violent crimes after it became a shall-issue state—very nearly 80% over the first two years.
Citation please? I spent some time googling to validate this and the best I could find was a book written about gun laws that cited a drop in murders for Oralndo that was near 80%. Other than that all the results I saw that showed violent crime statistics for Florida showed no drop in violent crimes anywhere near what you claim.
I like your blog, but how can you write a post about PDOOMA statistics and then include one such as this?
Crime stats for Florida showing the years after 1987 when shall issue was passed
Posted by on 11/27/2009 at 03:01 PMExcuse me, Vogon, but you are aware that not everything is freely available on the Web, aren’t you?
But you have at least one point: the statement should read ”certain violent crimes.” I shall correct it.
Posted by Francis W. Porretto on 11/27/2009 at 04:56 PMYou mentioned A.C. That’s enough stop me from reading anymore.
Posted by Charles Pergiel on 12/12/2009 at 08:32 PMOh, really, Charles? Well, tough shit, nancy-boy. Miss Coulter is worth a thousand of you. Take your excessively delicate sensibilities to some nice, lace-panty-friendly site where they won’t be so easily bruised.
Posted by Francis W. Porretto on 12/12/2009 at 08:38 PM
Comment Form
Commenting is not available in this weblog entry.
