Statistician-blogger Nate Silver says polling data from Research 2000 was likely too good to be true.
The research company was hired by Des Moines CBS affiliate KCCI-TV to examine Iowa races, most recently to gauge support in U.S. Sen. Chuck Grassley‘s re-election bid. On Tuesday, liberal blog Daily Kos released a report which suggests Research 2000 may have fabricated poll results.
In an e-mail posted on Silver’s blog, FiveThirtyEight, he wrote:
The age breakdowns in Research 2000′s numbers are almost always close to “perfect” — in 20 out of 20 cases, for instance, the Democrat gets a lower vote share from among 30-44-year-olds than among 18-29-year-olds. [Public Policy Polling]‘s data, on the other hand, is *much* messier — which is what I think we should expect when comparing small subsamples, particularly subsamples of lots of different races that are subject to different demographic patterns.
And during the 2008 president election season, Silver said there wasn’t enough variation in Research 2000′s data. “At no point, for instance, in the two months that they published daily results did Obama’s vote share fluctuate by more than a net of 2 points from day to day.” The accuracy of that little fluctuation is “extremely unlikely,” Silver said.
Here in Iowa, Research 2000 data indicated Democratic Senate challenger Roxanne Conlin was within striking distance of 30-year incumbent Grassley. In May, a KCCI-commissioned Research 2000 study suggested just 9 percentage points separated Grassley and Conlin. Data like that heavily influenced Conlin’s campaign rhetoric as she billed herself as the only primary candidate capable of winning in November.
However, other polling outlets see Grassley’s seat as relatively safe. Recent Rasmussen data gave Grassley a double-digit lead and more than 50 percent support in a match-up with Conlin.