Zogby Poll: Kerry back in Electorial Vote Lead....

[xyboymil]

Kerry Gains Momentum in the Battleground States - New Zogby Interactive Presidential Battleground Poll Reveals



Democratic Presidential candidate John Kerry did some good for his campaign by choosing John Edwards, a U.S. senator from North Carolina, as his running mate, the latest edition of the Zogby Interactive collection of polls in battleground states shows.

Mr. Kerry has regained an advantage in the Electoral College race for the White House in the latest round of simultaneous polling in 16 key states in the 2004 Presidential election, which began as Mr. Kerry was announcing his selection during a rally in Pittsburgh.

A close call yet...but based on the numbers, Kerry now would lead 322-205 over Bush.

http://www.zogby.com/news/ReadNews.dbm?ID=842

[calimary]

1. VERY telling that, yesterday, when they were parsing poll numbers on CNN,

(or maybe it was the day before),
Bill "American Enterprise Institute" Schneider spent most of his time talking about how Edwards hadn't made a dent in Kerry's ratings in Edwards' own home state. Well, okay, maybe there are still too many knuckledraggers there. And the anchor (I forget who) kept trying to steer him back to the overalls that showed Edwards HAD INDEED made a difference. He kept wanting to focus on the single state effect. They'll say ANYTHING.

[Kholst]

2. I don't buy Zogby's technique...

But his results aren't far from rasmussenreports.com

First, Zogby counts leads that aren't statistically significant as wins in this tally.

Second, and more importantly, his sample is biased. The sample is only those people who visitted his web site, signed up to receive emails from him, and responded to an email asking for participants. I have trouble believing that this makes for an accurate cross-section of the country.

As I said, though, Rasmussen has Kerry with a lead in the electoral college as well. I'm still lookign into their techniques.

[Kolesar]

4. Good catch on the polling techniques ... eom

[Robin Hood]

5. Ehhhh, Wrong!!! Thanks for playing.

"In our national telephone surveys, the gap in the south closed from Bush 53% to Kerry 35% in early June. But now it is Bush 50% to Kerry 40%. If these numbers hold in the south, the President's game play will be thrown off as they will at least have to spend more time and resources in the previously thought safe region."

Bwahahahahaha!

[Kholst]

10. ummm... that's not the poll we're talking about...

Zogby does traditional (telephone) polling as well, but his electoral college state-by-state polls are based on what he calls "interactive polls"

right on the page linked to in the title post:
"Zogby Interactive respondents were invited to participate from panels of likely voters who have agreed to take part in online surveys. Respondents followed instructions leading them to the survey located on secure servers at Zogby Interactive headquarters in Utica, New York. All surveys were completed between Tuesday, July 6 through Saturday, July 10. Slight weights were applied to region, party, age, race, religion, and gender to more accurately reflect the votingpopulation in the United States."

[Reciprocity]

12. Yes I was one of them.

All surveys were completed between Tuesday, July 6 through Saturday, July 10.

[Gman]

7. Zogby's interactive poll numbers

are usually pretty close to the telephone call numbers. Zogby is pioneering the online technique and takes into consideration the inherent problems with the interactive method. He appears to be refining it as he goes and its starting to look pretty accurate.

Again, don't forget Zogby was the only one that called '00 for Gore with his polls so he definitely knows what he's doing and how to remove any inherent biases. Its not perfect but its getting there. With the proliferation of cell phones and averseness to solicitiation calls at home, interactive is the way of the future.

[Kholst]

11. I don't question Zogby's skills

I think he's probably doing interactive polling better than anyone and will get better at it still, but there are too many filters and the technique is too new not to treat the results with a heavy dose of skepticism. You take the overall population, crop it to only those with internet, further crop it to only those who visit a specific polling site, further crop it to only those who sign up, further crop it to those that Zogby invited to participate, and finally crop it to those who respond to the invitation... Except for the second to last, I have no doubt that each step skews the sample from the general population. I know Zogby adjusts the results based on demographics, but the more adjustment that needs to be made, the wider the margin of error introduced.

His results are interesting and maybe useful (especially considering how few traditional polls even attempt a state-by-state poll), but should be viewed with the awareness that the technique is far from perfect.

[Beetwasher]

3. Damn! TN is tied???

That's awesome. The trend is clear. Kerry is strengthening his position in the battleground states that leaned Kerry and opening up new one's that weren't battlegrounds before, like TN and VA!

[xyboymil]

6. Agreed...

The biggest nightmare for GW Bush is that he loses ground in the south and has to spend campaign money/time to get states he thought were his.

If that happens, he will lose the Midwest. YES!! :toast:

[alfredo]

8. Not good enough. We have to get bush under 200 EV's.

dig in folks, we want this to be a clear message to bush and his Junta.

[soup]

9. There would have been one more Florida respondent if

Yahoo hadn't put my invitation from Zogby to participate in that particular poll in my SPAM file. :grr: