I would doubt the accuracy of this poll.
First, of all the major polling agency, Zogby has often been near the bottom in terms of accuracy. From my experience, Zogby often happens to come out with the "surprising" poll--the one no one else agrees with. In most cases, this out-of-the-norm prediction turns out to be incorrect.
Another thing that makes me question the poll results is this internal statistic, in reference to a question as to whether the choice of Palin will help or hurt:
Among Republicans, the choice was a big hit - as 87% said it would help, and just 3% said it would hurt.Who exactly is Zogby polling--Sarah Palin's immediate family? (Actually Palin's husband isn't a registered Republican, so even that doesn't explain it.) Just a quick survey of conservative editorials and blogs show a percentage far greater than 3% that question the choice. Sure, it was a popular one--but only 3% that think it a poor once seems far too low.
The other major polls such as Rasmussen and Gallop are 3-day tracking polls, so it will take a couple of days before the data here is all post-Palin. Currently Rasmussen shows the spread at +3% for Obama, but they never showed a convention bounce for him greater than 4%. Gallup still shows Obama up by 8, but the data from this poll includes no post-Palin information.