ARG Iowa poll: Clinton 30%, Obama 24%, Edwards 19%

[maximusveritas]

http://americanresearchgroup.com/pres08/iadem8-710.html

Not too much of a change from previous results, though it seems to confirm that Obama and Clinton have surged past Edwards.

Sample Size: 600 completed telephone interviews among a random sample of likely Democratic caucus goers living in Iowa (496 Democrats and 104 no party (independent) voters).

Sample Dates: September 26-29, 2007

Margin of Error: ± 4 percentage points, 95% of the time, on questions where opinion is evenly split.

[jefferson_dem]

1. ARG tends to offer funky numbers from time to time.

Nonetheless, we'll file this one away with the others...that show Edwards in big trouble in Iowa.

[wyldwolf]

4. "Funky numbers" when Obama isn't in first.

[jefferson_dem]

6. LOL! There you are...making shit up again.

Do a search and you'll see that I, like others, have routinely questioned ARG's reputability and methodology.

EDIT: Here's a link for you, wyld. Educate yourself. http://www.pollster.com/blogs/the_effect_of_arg_polling_on_i.php

[wyldwolf]

9. I've never 'made shit up,' but in regards to ARG...

... all I see if people trying to figure out why there results are sometimes different. Here is what we read from you link and related links:

American Research Group (ARG) does a large amount of state primary polling and is therefore potentially influential in estimating candidate support because they contribute more polls than most other organizations.

and...

Following up on yesterday's post, in which I speculated - wrongly, as it turns out -- about the incidence of eligible adults selected by the American Research Group (ARG) as likely caucus goers for their most recent surveys of Democrats and Republicans in Iowa.

They do MORE polls. People trying to discredit them have been WRONG in their theories.

[jefferson_dem]

11. Leaving aside the issue of whether or not you make shit up...

The fact that ARG does more polls does not make those who question their methods and results wrong. We'll only know who's right there after the votes are cast, eh.

The link i provided demonstrates how they routinely overestimate Hillary's support, as compared to other firms. There's more "out there" on the unreliability of ARG if you were really interested.

At any rate, your initial BS point was that I only criticized ARG's numbers this time because Obama's not leading.

Cheers. :toast:

[wyldwolf]

13. Why leave it aside?

The fact that ARG does more polls does not make those who question their methods and results wrong.

But the fact that those questioning their methods backtracked does. It certainly makes those who question the polls, then say "well, we were wrong about them" look silly.

The link i provided demonstrates how they routinely overestimate Hillary's support, as compared to other firms.

No, the link you provided demonstrated how Hillary routinely leads in their Iowa poll. It doesn't say they overestimate her support. In fact, the link shows their methods are reputable as the writer at your link twice had to admit their "theories" about ARG were wrong.

And, in fact, you've not been critical of polling showing Obama leading, have you? I mean, in this thread:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=3562940&mesg_id=3562940

Just 200 were surveyed... showing Obama leading. Did you question that? No.

So, not really bullshit.

[Cameron27]

2. It sure looks that way,

Obama & Clinton are both surging ahead on poll numbers and money. It's still early though, anything can happen.

[Colobo]

3. It's Clinton vs. Obama in Iowa then!

[emilyg]

5. Exciting!!!!

[Colobo]

7. Oh yes indeed! I wish they could have a 90 minute debate for themselves.

[emilyg]

8. So do I. That would be riveting.

[geek tragedy]

10. Clinton vs. Obama.

Pick sides folks, because those be your choices.

[rwheeler31]

14. crap media

[creeksneakers2]

12. This result

is more in line with other polls.

[wyldwolf]

15. yes, it is. And that's important to note.