Replies

Darth, the lunacy of the Zogby poll just dawned on me this week when I was contacted to participate in this interactive poll. This is important because Z. had contacted my e-mail address before and I deleted it without even looking at it. Now, you know the numbers better than I, but doesn't Zogby's interactive "battleground" poll use something like 1300 respondents from 10-15 battleground states? (I forget the specifics). What this means, though, is that fewer than 100 people per state are responding, and that means that when I added my little vote for Bush, it added something like 1/800, or 1.5% of the total!!! Imagine, now, 2-3 other Freepers did the same thing. You could get a shift of 4.5% instantly! It's bogus!

I would also expect that Zogby doesn't use all the "votes" he receives, but adjusts them based on making sure he has certain demographics represented that conform to his estimate of how people will vote in that state. Dales has a better discussion of the problems with Zogby Interactive, and has pointed out that recent traditional polls disagree with the ZI results.

I don't use the Zogby Interactive numbers in my analysis.