Insert Zogby's state poll numbers into Electoral-vote.com today and ...

[TruthIsAll]

... you will see that Kerry has over 320 EV.

Survey-USA?
Gallup?
Strategic Vision?
Mason-Dixon?


Trust Zogby. Trust ARG.

Don't be bamboozled by Repub pollsters.

http://www.geocities.com/electionmodel /

[DeminDC]

1. Question

The Zogby poll is an interactive survey. Why should that be considered more scientific than the telephone-based surveys? It is easy enough for us to infiltrate the Zogby sample.

[TruthIsAll]

3. Zogby is a pro. Along with Harris, he came closest in 2000.

He knows what he is doing.

He would not be using polling methodology if he were not confident that it was sound.

The fact is, all the Repub pollsters are biased - by definition.

Zogby works with Reuters and the Wall Street Journal.

Nuff' said.

[Dancing_Dave]

5. Why Zogby's New Method is Supposed Improvement

Zogby did telephone polling for years, and still does some, but John Zogby feels that telephone polls have an unacceptable bias in this election. The reason is that cell phone numbers are never used. They don't come in reliable area coded lists that polsters know what to do with. But most college students these days have ONLY cell phones, and no residence phones of their own. This demographic is known from University-based polls to be anti-Bush.

Of course, there's some other demographics that prefer cell phones exclusively. Have you met any Drug Dealers for Bush? As far as I can see, the only drug dealers likely to vote for Bush are the ones who really work for the CIA! Or at least have some kind of a buisness arrangement with the Bush family....

Zogby's internet method is supposed to send the interactive poll invitations in some statistically balanced method. It's not just a poll he puts up at some site, which would obviously be susceptible to a lot of partisan bias, and perhaps some technical manipulation as well. How do they KNOW each person only votes only once in those on-site polls? Anyone with access to a few computers and perhaps a little technical cleverness can find a way to enter multiple votes!

Zogby thinks he's found a way around these problems. I'd trust his results a bit more than most "lies, damn lies and statistics".

[Dancing_Dave]

2. I don't trust any polls too far....

All of them seem to do polls with too small sample size, etc.(statistical issues), and then there's the fact that people don't always do the same in a private election booth as they do when talking to a stranger who calls on the phone asking their opinion(psychological issues).

But Zogby is certainly one of the best. And the model you link to does seem to have a quite sound statistics/probability method. It's probably good news.:)

[TruthIsAll]

4. The sample sizes vary, but are good for 3.1% national and 4% state MoE's.

I like the Economist poll, which samples over 2000.

In any case, by calculating the average of a group of polls, we effectively lower the MoE. That's what I do in the model for the 9-poll and 18-poll groups.

It's too expensive to do poll more than 1000 respondents.
So I do it for them.

Averaging 9 polls of 1000 indivdiuals is equivalent to a single poll of 9000. This lowers the MoE from 3.10% to 1.03%.

tia

[question everything]

6. CNN worries me, saying Bush has 301 E.V.

http://www.cnn.com/2004/ALLPOLITICS/10/08/electoral.map/index.html

f the election were held today, Bush likely would win 301 electoral votes to Kerry's 237, according to a new CNN survey based on state polling as well as interviews with campaign aides and independent analysts. A candidate wins the election with at least 270 electoral votes, regardless of the popular vote.

[TruthIsAll]

7. If CNN said so, it must be true. Not.

tia

[DaveinMD]

8. I think

if the election was today, Kerry would have around 280. I think he'll end up with over 300.

[IIgnoreNobody]

9. I question CNN's numbers

[question everything]

10. Thanks. I appreciate it but would like to know how they got them

and how we dispute them.

I don't even know CNN track record

[Quixote1818]

11. By combining all the polls Kerry is clear leader. Take a look:

http://www.pollbooth.org/august20.shtml

http://www.race2004.net/index.php

[DuaneBidoux]

12. But Zogby looks very interesting for several reasons...

First and foremost, he consistently comes up with the largest percentage of undecided votes. Historically, undecideds break against the incumbent 2 to 1. At this point, if you're undecided it's obvious you don't like shrub but are unsure that challenger would be better.

Also, and in this way I think Kerry's debate strategy was thought out, to appeal to the base because Democrats are still committing to Kerry in far lower percentages that Repugs are to shrub (93% I think). If Kerry could get 90% of registered Dem's (vs. I think low 80s as Zogby now says) then it would not be close.

I was still a little disappointed by the debate, not by Kerry's performance, which was excellent. But it was intelligent, and thoughtful, and that doesn't necessarily do it. I was hoping Bush would seem as incompetent as the first debate, and he definitely didn't seem that incompetent. A repeat Bush performance from the first one would have made it much easier, that's for damn sure!