So many polls so little time: Iowa

Cross-posted at Daily Kos

I never dismiss a poll out of hand.  I think every poll has something to tell us.  The results of the Des Moines Register poll have been widely noted here and throughout the blogosphere.  Obama has a seven point lead:

Obama 32
Clinton 25
Edwards 24

This poll, the Register reports, has a +/-3.5% margin of error.  The model, according the The Washington Post, assumes that more than half of Hillary's and Edwards's support will be from first time caucus goers and that 72% of Obama's will come from that group.

Another few hours, and two more polls arrive this morning.  CNN suggests that the race looks this way:

Clinton 33
Obama 31
Edwards 22

This is a telephone conducted on December 26 to December 30. There were 482 likely Democratic caucus-goers interviewed for the survey.  It had a +/-4.5 margin error, suggesting a stalemate between Hillary and Obama.

The Reuter/Zogby tracking poll is out this morning, too:

Clinton 30
Obama 26
Edwards 25

According to Reuters:

The poll of 925 likely Democratic caucus-goers . . . was taken Friday through Monday and has a margin of error of 3.3 percentage points for each party.

So we wait till Thursday.



Display:


Re: So many polls so little time: Iowa (2.00 / 3)

Too much data!  I am predicting Hillary, but if you give me four buck, I'll be able to buy a venti latte at Starbucks.


Our long national nightmare is over.
by Beltway Dem on Tue Jan 01, 2008 at 07:50:17 AM EST

You're funny, Beltway Dem! (2.00 / 1)

I am predicting Hillary, but if you give me four buck, I'll be able to buy a venti latte at Starbucks.

I guess that's ALWAYS one safe bet! But don't worry, Hillary will do just fine on Thursday. All she needs to do is keep working hard reaching out to voters, and let Obama and Edwards fight over this crazy DMR poll. While those two fight and drag each other down, she may very well end up on top! ;-)


Want to defend marriage equality in Maine? Ask me how!
by atdleft on Tue Jan 01, 2008 at 10:45:38 AM EST
[ Parent ]

Re: You're funny, Beltway Dem! (2.00 / 2)

Hey, atd:  it's 2008.  Wow.  Time does fly.  Happy new year!

markjay has a much better diary about all of this at the top of the recommended list.


Our long national nightmare is over.
by Beltway Dem on Tue Jan 01, 2008 at 11:02:10 AM EST
[ Parent ]

And I like your diary, too... (none / 0)

'Cuz you point out all the new polls, and you show just how DMR is so out of line with all the other Iowa polls. And as much as I don't like to say this about a poll that's usually so respected, I think it's an outlier. I don't see how that many Independents and Republicans will attend a Democratic Caucus. It just makes no sense.

I have a feeling CNN, ARG, Rasmussen, and Zogby (gasp!) may be showing a more accurate picture here. While more Indies may show up this year, I don't think they will make up 40% of caucusgoers. While turnout may increase, I don't think their model is completely accurate.

But hey, only time will tell who's ultimately right... I'll bet on Beltway Dem! ;-)


Want to defend marriage equality in Maine? Ask me how!
by atdleft on Tue Jan 01, 2008 at 12:16:38 PM EST
[ Parent ]

Re: So many polls so little time: Iowa (2.00 / 3)

Agreed Beltway. But we are talking about the DMR. However, the DMR maybe wrong. As you said, let's wait until Thursday.

Side note: Obama's been on the attack the last week or so, which signaled that his internals were bad. Now the DMR says otherwise. Why would he be going negative? This seems odd to me. I still think it is a 3 man/woman race. Can't wait until Thursday.


by lonnette33 on Tue Jan 01, 2008 at 07:51:27 AM EST

If even Obama's internal polling... (2.00 / 3)

Doesn't predict these many Indies showing up to vote...

http://marcambinder.theatlantic.com/arch ives/2007/12/my_one_question_about_the_r egi.php

Then I think there may be something wrong with the DMR poll. As we talked about in another thread, this just may be the year the DMR poll turns out to be wrong. Our friends in Iowa just need to ignore this and keep working hard, and Hill can win this thing. :-)


Want to defend marriage equality in Maine? Ask me how!
by atdleft on Tue Jan 01, 2008 at 10:52:04 AM EST
[ Parent ]

Re: If even Obama's internal polling... (2.00 / 1)

Agreed!


by lonnette33 on Tue Jan 01, 2008 at 12:13:24 PM EST
[ Parent ]

cargo cult re: independents (2.00 / 1)

the weird thing is that the obama supporters i'm hearing say it's a self-fulfilling prophecy- that the poll will affect the outcome by encouraging obama supporters to stick with obama. why would they say that if they believed the poll?  it sounds like a tacit admission that they know it's high. i didn't believe that 14 point hillary lead from ARG last week, but i didn't kick it out of bed, either ;-)

the more independents you include in your sample, the more you inflate their influence, because they show up less reliably.  but you've got to weight them equally with people who have attended the last six caucuses, because all you have is self-report for intent to caucus. a "yes" is a "yes". and with first-time caucusers, it's all theoretical. they HAVE no history of caucusing. if you ask, "did you caucus in 2004?", that's real behavior you'd have a firmer base to use as a
prediction.

as an example, both mrs campskunk and my slacker dopehead nephew have stated their intent to take care of me when i'm 90 and need diapers. should i weight their statement equally?


by campskunk on Tue Jan 01, 2008 at 11:06:14 AM EST

Re: So many polls so little time: Iowa (none / 0)

I know I'm going against the polls here but my prediction is

Edwards

Clinton

Obama

basely solely on the reliance on first time voters for each candidate. For years and years, I've heard how so and so was going to win because first time voters were going to show up. Well, they largely never do and I have seen no evidence that makes this year any different than other years.


No longer a Democrat, now proudly an independent voter!
by Ga6thDem on Tue Jan 01, 2008 at 12:45:15 PM EST

Re: So many polls so little time: Iowa (none / 0)

Obama could win in New Hampshire if there is a coordinated attack by Republicans against Hillary.  Registration is possible at the night of the caucuses, and so if a lot of Republicsns and Independents show up at the Iowa caucuses just to make sure Hillary doesn't win, the results could be skewed by those efforts.


by FarWest on Tue Jan 01, 2008 at 01:30:00 PM EST

Re: So many polls so little time: Iowa (none / 0)

Obama could win in Iowa if there is a coordinated attack by Republicans against Hillary.  Registration is possible at the night of the caucuses, and so if a lot of Republicsns and Independents show up at the Iowa caucuses just to make sure Hillary doesn't win, the results could be skewed by those efforts.


by FarWest on Tue Jan 01, 2008 at 01:30:19 PM EST


You are not logged in.

In order to post a comment, you must be logged in. If you have a member account, please log in to comment.

If not, you can make an account right here. It's quick and free.