Pollster Strategic Vision LLC is in a bit of trouble after a failure of transparency in regards to their survey methods. Folks at 538.com took it to a new level when they found seemingly non-random non-uniform distributions of trailing digits (e.g. 53-47 McCain-Obama). In a recent post, they use Fourier analysis to determine the likelihood of the discrepancies to be the result of a systematic methodological error. The result:
The Strategic Vision result, on the other hand, or something more extreme, would occur by chance with probability only 0.00019. That’s not as low a p-value as the results obtained without filtering the non-uniform components, but it’s still very low -- less than one chance in 5000 to have occurred by chance alone.
In other words, statistics demonstrate that Strategic Vision has been cooking their books. "Nerdgasm," indeed.
On a slightly related note, Tom Levenson (director of the science writing grad program at MIT) pretty thoroughly dissects Megan McArdle's entire mode of thinking. McMegan inexplicably shares blogspace with TNC and Sully over at The Atlantic. She is known for blogging as "Jane Galt" and lying. Levenson establishes McArdle as a truly professional cherry picker of knowledge, not knowing--or not caring--enough to think critically about the data sources she uses. If you can get through it (4,000 some words), you get a great analysis of how the phrase "conservative think tank" (or liberal think tank, for that matter) is irrevocably oxymoronic.
Shorter version at BJ.
Chris Mooney (different Chris Mooney) has the latest debunking of climate myths. Unfortunately this is something we will never get enough of. 27%!
Posts
0 comments:
Post a Comment