New USA Today/Gallup: Clinton 50, Obama 22, Edwards 15, None/Other 6

[rinsd]

4. Which of the following candidates you would be most likely to support for the Democratic nomination for president in 2008?

BASED ON 508 DEMOCRATS AND DEMOCRATIC LEANERS

Without Al Gore

  	       Clinton Obama Edwards Richardson Biden Kucinich Dodd  Gravel  Other  None/ no opinion
2007 Nov 2-4 	50 	22 	15 	4 	1 	1 	1 	* 	1 	5
2007 Oct 12-14 	50 	21 	13 	4 	3 	1 	* 	* 	1 	6
2007 Oct 4-7 	47 	26 	11 	4 	2 	1 	1 	* 	1 	5
2007 Sep 14-16 	47 	25 	11 	5 	3 	2 	1 	1 	* 	4
2007 Sep 7-8 	45 	24 	16 	2 	3 	2 	1 	1 	2 	5

5. Suppose the choice for the Democratic presidential nomination narrows down to Hillary Clinton and Barack Obama. Which one would you prefer the Democratic Party nominate for president?

  	     Clinton   Obama 	No opinion
2007 Nov 2-4 	60 	33 	7

Plenty of goodies for poll junkies at the link http://www.usatoday.com/news/polls/tables/live/2007-11-06-politics.htm

For results based on the total sample of national adults, one can say with 95% confidence that the margin of sampling error is ±3 percentage points.

For results based on the sample of 430 Republicans or Republican leaners, the maximum margin of sampling error is ±5 percentage points.

For results based on the sample of 508 Democrats or Democratic leaners, the maximum margin of sampling error is ±5 percentage points.

[Beaverhausen]

1. polling 508 democrats means nothing

How many democratic voters are there nationally? polling a sample that small is pretty useless if you ask me. Not that you did...

[rinsd]

2. Most of the primary polls have anywhere from 400 to 700 voters in the sample.

See for yourself - http://www.pollingreport.com/wh08dem.htm

I also added the MOE's which I should have done in the initial OP.

[Kurt_and_Hunter]

3. The usefulness of polling 508, 10000 or 12 is objectively determinable

The math tells us precisely how useful a sample of 508 is, so there's no need to guess.


For results based on the total sample of national adults, one can say with 95% confidence that the margin of sampling error is ±3 percentage points.

For results based on the sample of 430 Republicans or Republican leaners, the maximum margin of sampling error is ±5 percentage points.

For results based on the sample of 508 Democrats or Democratic leaners, the maximum margin of sampling error is ±5 percentage points.

[magnolia]

4. Polls...

...polls always mean nothing when one doesn't like the results. Funny that.

They do it at freerepublic all the time. If they are in our favor, they are crap...if in their favor they are whooping it up.

You either trust polls or not. If you don't then why bother to look at them and comment on them.

[Carrieyazel]

5. National polling is essentially meaningless, but even it weren't:

She's not at 50% among Dems. That figure is not to be believed.

[Alamom]

6. k&r