Patrick Ruffini looks at the kinds of polls Research 2000 said it was providing Daily Kos, and concludes most pollsters would have charged an exorbitant amount for the amount of time it required to conduct their polls. The infamous Republicans-are-lunatics one, for example, would have cost at least $100,000 for most pollsters with the sample size, number of rejected respondents, and so on. Ruffini notes that if the price was significantly less, “anyone with a rudimentary understanding of polling would have known you can’t do a poll like this for that amount of money.”
Yesterday I noted that Research 2000 was the lone pollster I could find that listed a P.O. box as its mailing address.
Another oddity that Campaign Spot reader Patrick noticed; their web site is www.Research2000.us, not Research2000.com.
The latter URL currently goes to no web site, but is apparently owned by a Research 2000 Inc., based in Cleveland, Ohio. That company owns the domain name through 2011. At least as of 1996, that company appears to be doing something with “electronic structure, surface dipole properties, and work functions of scandate surfaces,” and is not a polling company.
Another oddity: There is no statement, response, or even acknowledgment of the Kos controversy on the company’s website; the front-and-center is a poll from April 11. If someone had just accused your company of falsifying poll results, wouldn’t you have your statement/denial/counter-accusation front and center?
Ironically, the Google News feed on the site links to several stories about the controversy, including the post in which the statisticians studied the results and told Kos, “We do not know exactly how the weekly R2K results were created, but we are confident they could not accurately describe random polls.”
I have worked with surveys for over a decade and I agree: the amount of polling that was done should have added up to a much larger amount than Kos is suing for. Especially considering the boast that they did all in-person phone interviews (now you're talking at least $20 CPI for data collection, not including analysis and everything else.)
I wonder if Kos ever bothered to get a competing bid? Ha!
That statistician's remarks make it sound like the were basically faking all their data -- which is consistent with their cost structure.
Reply to this commentLinkReport AbuseToo funny. Pass the popcorn.
Reply to this commentLinkReport Abuse