A couple more notes about this morning’s Quinnipiac poll, putting Democrat Jon Corzine up by 5 percentage points. Quinnipiac is usually a fine pollster, but a couple of factors about this one are starting to emit a bit of an odor.
Mark Impomeni heard back from Quinnipiac that the sample splits 49 percent Democrat, 27 percent Republican, 23 percent Independent. [See UPDATE: Now Quinnipiac says the sample splits 40 percent Democratic, 25 percent Republican, 29 percent Independent.]
As noted earlier, on Election Day 2008, CNN’s exit poll had it at 44 percent Democrat, 28 percent Republican, 28 percent Independent. So we’re to think that the Democrats, with Jon Corzine at the top of the ticket, will make up 5 percent more of the electorate than the did in 2008, with Barack Obama at the top of the ticket?
Another reader noted that in Quinnipiac’s last poll in New Jersey, Christie led among independents from 42-31; now in this most recently poll, he leads 45-30; yet overall, Christie has lost his 41-40 lead to trail 43-38.
Also, note this line in the release: “From October 20 – 26, Quinnipiac University surveyed 1,267 New Jersey likely voters, with a margin of error of +/- 2.8 percentage points.” A poll taken over seven days? Including a weekend? That seems like a minor problem…
UPDATE: Impomeni says Quinnipiac now says the sample splits 40 percent Democratic, 25 percent Republican, 29 percent Independent. That’s interesting, because a Republican in New Jersey had told me his back-of-the-envelope math was that Democrats had to split somewhere between 45 percent and 48 percent of the sample.
I’m left wondering where the other 6 percent went. Say, good people at Quinnipiac, how about responding to my e-mail? Or even better, putting this information out in the original release?