ARG Polls Iowa, New Hampshire and South Carolina

by Chris Bowers, Wed May 02, 2007 at 02:03:50 PM EST

New polls of the early states are out from ARG. The New Hampshire numbers are really, really strange, but the Iowa and South Carolina numbers look reasonable:

600 "likely voters / caucus goers", MoE 4
Candidate IA, 4/30 IA, 3/22 NH, 4/30 NH, 3/22 SC, 4/30 SC, 2/27
Clinton 23 34 37 36 36 36
Edwards 27 33 26 20 18 20
Obama 19 16 14 23 24 25
Biden 6 2 2 2 3 2
Richardson 5 1 3 2 1 1
Kucinich 2 1 2 1 3 1
Dodd 2 1 1 1 1 1
Other / Unsure 16 12 15 13 13 14

Clearly, the big "winner" after the first debate was anyone but Clinton, Obama and Edwards. In Iowa, Biden, Dodd, Kucinich, Richardson went from a combined 5% in March to a combined 15% in late April, while Clinton, Edwards and Obama dropped from 83% to 69%. Overall, that is a 24% swing between the two groups. The field made smaller gains in New Hampshire and South Carolina, but they were gains none the less. Also, Richardson has not been able to establish himself as "along in the second tier," where right now he clearly shares company with Biden. I didn't list Gravel because he did not reach 1% anywhere. That might be a reasonable benchmark for debate attendance: a candidate must have 1% both nationally and in the state where the debate is being held.

But there is something very strange about the New Hampshire poll. Obama is supposedly a distant third in New Hampshire now? He is doing better in Iowa than New Hampshire? Edwards is further ahead of Obama in New Hampshire than he is in Iowa? That all seems very unlikely to me. Unfortunately, there have been precious few Iowa and New Hampshire polls over the last month, so right now we have no other results to which we can compare these numbers. Since ARG has proven consistently wacky, I am going to assume the New Hampshire poll is an outlier until other polls appear to confirm it.




Tags: Iowa, New Hampshire, South Carolina, President 2008, polls, Democrats (all tags)

Display:
Permalink :: 53 Comments :: Post a Comment


Obama down in NH (none / 0)

I don't think Obama's post-debate drop in NH is strange at all.

First, I thought Obama's debate performance was weak in a number of areas.  Mostly, his rambling about Katrina and first-responders after an Al-Qaeda attack doesn't play in NH; home to groups like "Daughters of the American Revolution".

Second, Obama agreed to debate in NH only yesterday, and only after Clinton said she would join the debate.  

Lastly, it my understanding that Obama hasn't and wasn't planning on doing alot of campaigning there.  

I'm sure many in NH took these latter two positions as somewhat of a snub.


by ChicagoDude on Wed May 02, 2007 at 02:19:09 PM EST

Re: Obama down in NH (1.50 / 2)

are you paid by Hillary?


Obama! because 51% isn't enough!
by nevadadem on Wed May 02, 2007 at 02:20:12 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

Ah yes, instantly accusing someone of a hidden agenda. Seems to be a habit of many around here these days.
by Chris Bowers on Wed May 02, 2007 at 02:23:44 PM EST
[ Parent ]

Re: Edwards UP in NH (none / 0)

Edwards took poll points from Obama, it appears. Perhaps Edwards message worked in NH and Obama's didn't.

Iowa is a different story. Hillary tanked in Iowa and Edwards moved up.

It looks to me like Edwards was the only one of the top three to move up significantly after the debates, but he didn't do well in South Carolina. Maybe his message has become too progressive for South Carolina.


Children, have you any fish?
by FishOutofWater on Wed May 02, 2007 at 02:29:10 PM EST
[ Parent ]

Re: Edwards was 1st in Iowa (none / 0)

Even though his poll numbers dropped, he moved ahead of Hillary who plummeted. Hillary lost points to Richardson, Biden and other candidates who were struggling for exposure.

We shouldn't take these numbers too seriously.


Children, have you any fish?
by FishOutofWater on Wed May 02, 2007 at 02:33:30 PM EST
[ Parent ]

Re: Edwards was 1st in Iowa (none / 0)

I wouldn't really take any polls seriously this far out... they really mean very little.


by yitbos96bb on Wed May 02, 2007 at 03:05:55 PM EST
[ Parent ]

Re: Edwards was 1st in Iowa (none / 0)

I agree. It's very difficult to screen for primary voters this far out, and for caucus goers it's even worse.

Plus we're dealing with very small sample sizes, so a few people having a bad day, or a inaccurate screen, can swing your numbers.  Most of the changes could be random swings without statistical meaning.


by dwightmc on Wed May 02, 2007 at 03:23:11 PM EST
[ Parent ]

Re: Obama down in NH (3.00 / 1)

Those sound like the words of a paid Gravel operative, Chris.


by LPMandrake on Wed May 02, 2007 at 02:29:25 PM EST
[ Parent ]

Gravel (none / 0)

Soon to be an SNL favorite I am sure.


by dpANDREWS on Wed May 02, 2007 at 03:13:38 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

Hey, he actually "likes" Hillary Clinton, the icy princess of the frost, the wicked witch of the East.  He MUST be a paid operative.   :-)


by georgep on Wed May 02, 2007 at 03:00:27 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

No, that's not possible.  Everyone who likes Hillary, including a majority of registered Democrats, is on her campaign's payroll.


by lorax on Wed May 02, 2007 at 03:20:07 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

Or received Campaign Donations...

Sorry, that was snarky but I couldn't resist.


by yitbos96bb on Wed May 02, 2007 at 04:18:23 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

Don't be ridiculous. Not even Hillary's paid operatives like here. They hate what they're doing, but if they leave her campaign she'll kill their dogs.


Visit Forgotten Countries, my new foreign policy-based blog
by Englishlefty on Wed May 02, 2007 at 06:57:12 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

That's why I call her "Hitlerry"


by lorax on Wed May 02, 2007 at 08:52:18 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

Really? I assumed it was the moustache...


Visit Forgotten Countries, my new foreign policy-based blog
by Englishlefty on Fri May 04, 2007 at 12:58:40 PM EST
[ Parent ]

Obama supporters remind me of Republicans (1.00 / 0)

Many do anyway.

They remind me of wingnut supporters that have this ability to seperate their beliefs from reality.  Like when a preppy from CT becomes a cowboy from TX.   Anyone who believes otherwise is the enemy and must be attacked.

I think many Obama supporters create this narrative in their head about who Obama is and anyone who knocks it is instantly deemed the enemy.

My fear is that they will elevate this guy before they, or anyone else, really know enough about him and his abilities on the national state.  


by dpANDREWS on Wed May 02, 2007 at 03:12:37 PM EST
[ Parent ]

Re: Obama supporters remind me of Republicans (none / 0)

That's pretty rich coming from you.


by yitbos96bb on Wed May 02, 2007 at 04:18:46 PM EST
[ Parent ]

Am I begging for donations? (none / 0)

Look in the mirror.


by dpANDREWS on Wed May 02, 2007 at 10:35:21 PM EST
[ Parent ]

Re: Obama down in NH (none / 0)

I'd monitor his comments before rushing in to rescue.

Google (Andy Martin History). This same user name is on Huffington Post as well. It says even worse.


"I don't believe in this can't do, won't do, won't even try style of politics. Yes We Can!" ~ Barack Obama
by ObamaEdwards2008 on Wed May 02, 2007 at 04:21:52 PM EST
[ Parent ]

Some Iowa Polls, none that recent though (none / 0)

Iowa

March 19-31 , 2007
U. of Iowa

Results (Support in Caucus from Subsample of Democratic Caucus Goers)
John Edwards 34.2%
Hillary Clinton 28.5%
Barack Obama 19.3%
Undecided 12.5%

Results (Agree with "Blank is Electable")
John Edwards 89.0%
Barack Obama 86.6%
Hillary Clinton 76.5%

March 26, 2007
Zogby International

Results
Edwards 27%
Clinton 25%
Obama 23%
Undecided 15%



BlueNC - Progressive NC Politics
by Robert P on Wed May 02, 2007 at 04:44:14 PM EST
[ Parent ]

re (none / 0)

if Arg polls are to be believed every other pollster is and has been getting both the dem race and the GOP race wrong for months.......on the Perublican side ARG has McCain way ahead is the early states instead of Guliani, when I saw these #s yesterday I thought they were bad for Obama now after noticing the republican results I wonder how they are 20 poitns off everyone else.


Obama! because 51% isn't enough!
by nevadadem on Wed May 02, 2007 at 02:19:29 PM EST

Iowa doesn't make sense either (none / 0)

It's very very improbable that support for the Big Three would have dropped from a combined 83% down to 69% without any apparent cause.

I've been souring on ARG for awhile.   This goes back to after the CT primary last year when ARG was the only poll to show Lamont within 2% amongst the general electorate (which they did twice, IIRC) when everyone else was showing a 10-12% Lieberman lead.  (There've been some other anomalous results of theirs in between, but I can't remember what at the moment.)


by RT on Wed May 02, 2007 at 02:21:42 PM EST

another problem with the Iowa numbers (3.00 / 4)

Re-posting a comment I made in robliberal's diary yesterday:

One red flag for me in the Iowa numbers:

Sample Size: 600 completed telephone interviews among a random sample of likely Democratic caucus goers living in Iowa (475 Democrats and 125 no party (independent) voters).

On caucus night in 2004 we had a few people in my precinct change their registration that evening so they could participate in the Democratic caucus.

But anyone who believes that 1 in 5 Iowa caucus-goers on the Democratic side next January will be registered independents is crazy.

I don't know how including so many independents in the sample would skew the numbers, but it seems way off.


Join the Iowa progressive community at Bleeding Heartland.
by desmoinesdem on Wed May 02, 2007 at 02:26:00 PM EST
[ Parent ]

Re: another problem with the Iowa numbers (none / 0)

ARG'S NUMBERS WITH OUT INDEPENDENTS ARE:

EDWARDS 29
CLINTON 23
OBAMA 22


by BDM on Wed May 02, 2007 at 04:00:28 PM EST
[ Parent ]

ARG POLL OF IOWA INDEPENDENTS (none / 0)

Clinton     22%
Edwards   18%
Obama     10%
by ChicagoDude on Wed May 02, 2007 at 04:10:40 PM EST
[ Parent ]

Re: another problem with the Iowa numbers (none / 0)

OK, but presumably that was the case with both ARG polls.  And if so, then the inclusion of independents in both polls doesn't explain the change.  

If independents were only included in the more recent poll, then there's no comparing the two, and their heads are really up their asses.


by RT on Wed May 02, 2007 at 04:16:26 PM EST
[ Parent ]

Re: Iowa doesn't make sense either (none / 0)

Could the cause just be getting some airtime? Obviously that's dependent on how many people saw the debates (I freely admit to ignorance on that) but greater visibility is always going to lead to some movement in numbers.


Visit Forgotten Countries, my new foreign policy-based blog
by Englishlefty on Wed May 02, 2007 at 06:59:23 PM EST
[ Parent ]

Chris do you agree that the really strange GOP (3.00 / 0)

results make these polls even more suspect?


Obama! because 51% isn't enough!
by nevadadem on Wed May 02, 2007 at 02:22:06 PM EST

Re: Chris do you agree that the really strange GOP (none / 0)

We don't have any other IA and NH numbers to compare them too. I think they are suspect, but we will have to see when more polls come out.
by Chris Bowers on Wed May 02, 2007 at 02:23:12 PM EST
[ Parent ]

Re: Chris do you agree that the really strange GOP (none / 0)

Considering that Obamas base consists of College Educated Americans under 40 who are the least likely to be home on any given day, I'm so not concerned about these polls. I mean, who do you think is answering the phone?

"Eh Heh? speak up sonny. I can't hear hear ya. What? Heh? Clinton you say? Yeah I know who he is? Heh? President? Yeah, I remember him. I voted for Clinton. Am I going to vote? Sure, I'll vote. What time do I need to be there ? Heh? Who? 2008? Obama yeah we should find him. John Edwards? Yes I voted for Kerry Edwards last time. Yes. Ey?  Sounds nice. Heh? Okay, bye"


"I don't believe in this can't do, won't do, won't even try style of politics. Yes We Can!" ~ Barack Obama
by ObamaEdwards2008 on Wed May 02, 2007 at 04:16:22 PM EST
[ Parent ]

Re: Chris do you agree that the really strange GOP (none / 0)

Can do what?  Lost the Presidency one more time?  That's exactly what he'll do if he's nominated.  Look into his eyes.  He (Obama) is begging someone to tell him its okay he's not ready.  He's not ready.  


by wynter on Wed May 02, 2007 at 09:19:00 PM EST
[ Parent ]

Re: ARG Polls Iowa, New Hampshire and South Caroli (3.00 / 1)

Those NH numbers do seem like an outlier to me, too.

Here's what really caught my attention, though:

"85% of likely Democratic primary voters voting for Hillary Clinton, 71% of those voting for John Edwards, and 64% of those voting for Barak [sic] Obama say they do not believe the United States can win the war in Iraq."

It's just one question on one poll, but that's cause for concern (but maybe optimism as well) for both Edwards and Obama.  Clinton, the least anti-war candidate in the field, has the strongest anti-war support?  I think this may be proof that her campaign is effectively masking her positions while Edwards and Obama have not really broken through on this issue.  But does this also mean that she could be headed to a bigger fall if the positions of the major candidates start to become apparent to the primary electorate, rather than just the activist class?  


by LPMandrake on Wed May 02, 2007 at 02:27:49 PM EST

Obama's obvious ad (none / 0)

It's hard to believe but most dems that answer a telephone poll aren't really following this like we are, they probably barely know any of the candidates specific differences at this point. Obama's trump card is the obvious comparison ad of the 2002 vote at that point all the "low-info" voters will know who was pro war and who was not and Obama will have plenty of money to spend to get that message out.


Obama! because 51% isn't enough!
by nevadadem on Wed May 02, 2007 at 02:32:01 PM EST
[ Parent ]

NEW SOUTH CAROLINA POLL (none / 0)

A new South Carolina Poll seems to support ARG's:

Clinton    38%
Obama    21%
Edwards  17%

http://www.pollster.com/mystery_pollster /poll_branding_democratic_candi.php


by ChicagoDude on Wed May 02, 2007 at 02:58:13 PM EST

Re: NEW SOUTH CAROLINA POLL (none / 0)

Taken Way before the debates.


by yitbos96bb on Wed May 02, 2007 at 03:07:45 PM EST
[ Parent ]

Re: NEW SOUTH CAROLINA POLL (none / 0)

That is a massive jump in Clinton's numbers in South Carolina.   A similar jump can be observed when you look at polling in California and Florida.  Clinton went into blowout territory there recently, where before Obama was quite competetive in both California and Florida.

The "brand" discussion in that poll is also very valuable.  Chris Bowers should probably look at the findings and turn that into a diary.  It shows the strength and weaknesses of each candidate, which can help the candidates in their quest to improve their perceived weak areas.  Obviously, on balance, South Carolinan Democrats value Clinton's strong points over Obama and Edwards.        


by georgep on Wed May 02, 2007 at 03:08:12 PM EST
[ Parent ]

Re: NEW SOUTH CAROLINA POLL (none / 0)

I will concede thing nobody really knows who a likely primary voter at this at stage actaully is, WSJ has the most interestedfavoringObam38/31
and Rasmussen who's been very accurate has Obama winning right now, It seems like most of the fluctuations in the polls are due to the pollsters not anything really changing in the the race, which shows hllary slighly ahead but without much room for "growth" nationwide, as an Obama fan while ofcourse it looks good to take the lead the best position for him is slighly behind nationally going into Iowa and New hampshire so hillary is denied the "comeback" spin after losing that her Husband pulled of in 1992, Hillary's inevatablilty pushers like Mark Penn fail to realize that what "inevitability is good for has already failed, drying up other candidates money supply as Obama and even Edwards have plenty so now maybe she'll get more establishment endorsements but she was always gonna get those anyway, but now she is burdened with Dean's curse a bad showing in Iowa and/or New hampshire crushes her and makes any other state polling irrelevant because she'll have the "feel" of an unelectable loser because she was "supposedly inevitable".
Obama! because 51% isn't enough!
by nevadadem on Wed May 02, 2007 at 03:21:42 PM EST
[ Parent ]

Re: NEW SOUTH CAROLINA POLL (none / 0)

ARG uses a very lose screen for their likely voter model.They donot ask if the person is registered to vote or has voted in the last election.

They ask the following:

1. will you definitely vote in the 2008 primary or caucus.

2.definitely might may vote in the primary or caucus

3.definitely not vote in the primary or caucus

The question is which is their likely voter model number 1 only or both number 1 and 2.

IA resulys with out independents are as follows

Edwards 29
Clinton 23
Obama   22


by BDM on Wed May 02, 2007 at 03:52:30 PM EST
[ Parent ]

POLL OF IOWA INDEPENDENTS ARG (none / 0)

Clinton      22%
Edwards    18%
Obama      10%
by ChicagoDude on Wed May 02, 2007 at 04:13:35 PM EST
[ Parent ]

Re: POLL OF IOWA INDEPENDENTS ARG (none / 0)

What does that have to do with anything... if you are posting this follow up with something.


by yitbos96bb on Wed May 02, 2007 at 04:21:55 PM EST
[ Parent ]

Re: NEW SOUTH CAROLINA POLL (none / 0)

mOORE mETHOD POLL 4/14-4/19 cALIFORNIA

cLINTON 31
Obama 21
Edwards 19

Not such a big lead


by BDM on Wed May 02, 2007 at 04:05:37 PM EST
[ Parent ]

Re: NEW SOUTH CAROLINA POLL (none / 0)

BDM, gone are the days when you listed the SC "Insider Advantage" poll at the bottom of your posts.  Given that every other poll has shown a  Clinton lead (which appears to be growing, btw.,) that one obviously stunk to high heaven.

  Now you have to try to find the lower of the double-digit Clinton leads in the state to make a point.   :-)


by georgep on Wed May 02, 2007 at 04:44:08 PM EST
[ Parent ]

Why is this poll not in the MSM? Or talked about? (none / 0)

Until all these early states, IA, NH, SC and NV are polled on the regular, meaning to compare trends, then this poll is nothing.  Why is it that the MSM is touting the Rasmussen poll stating Obama is 2 pts ahead of Clinton, natioinally?  But realistically dead heat.  Because Rasmussen polls on a consistant basis.  This ARG poll, if you look at the questions, they did not even asked if you are a registered voter?  And it is ludacris to state that Obama is loosing support in SC, when thousands waited on Friday to hear him speak?  And he had almost 45K for the month of April.  It does not add up.  Lastly, you will know when a candidate is in trouble with the "money figures".  Q2 will be an indicator of how these candidates are pacing themselves, how the support is continuing, how many "new" people donated, ect.  That is one of the indications of how the support really is out there, for all these candidates.


"I want my voice to be read"
by icebergslim on Wed May 02, 2007 at 04:15:08 PM EST
[ Parent ]

Re: Why is (none / 0)

Rasmussen showed Bush with one of his best approval ratings of the year yesterday:  43%.   Harris has Bush at 28%, CBSNews/NY Times is at 30%, everybody else shows low 30s for Bush.  Yet, Rasmussen is held up as the gold standard by some, even though on the national polls numbers for Democrats they show the same discrepancy to all other polls (at least off by 8%, in many cases more.?  

State polls are more valuable than national polls.   You take an aggregate of all polls and come up with a relatively clear picture.  In the case of South Carolina, you have a picture emerging:  A once close race in the state has been blown wide open by Clinton.  You take the three recent polls and average them out:

ARG

      Clinton - 36%
       Obama - 24%
       Edwards - 18%

Chernoff Newman/MarketSearch:

      Clinton - 38%
       Obama - 21%
       Edwards - 17%

Moore Method:

       Clinton - 31%
        Obama - 21%
        Edwards - 19%

What becomes clear are two things:  Clinton has a strong, double-digit lead in South Carolina.  Plus, she has added to her previous lead in South Carolina, which at some point was rather slim.  

That growth trend has become apparent in other states as well, such as California and Florida.  It will be interesting to see if that trend continues.


by georgep on Wed May 02, 2007 at 05:11:10 PM EST
[ Parent ]

Re: Why is (none / 0)

Couple of points:

1. The Moore Method poll is of California not SC.

2. Rasmussen polls Bush JA every day and last week he had days when Bush was at 36-38 approval.
Yesterday Bush was at 41

3, Look at his record in which he was the most accurate Pollster in 2004.

4. Every month he adjusts his model of percent of dem's, repub.s and Independents.

Based on April's polling of 15,000 calls his model going forward is Dem's 36.5, Repub's 31 AND iNDEPENDENTS 32.4

rEPUBLICANS ARE AT THEIR LOWEST LEVEL SINCE 2004 AND dEM'S HAVE FALLEN FROM 38 TO 36.5 SINCE THEY HAVE TAKEN POWER.

My take is that the race is very fluid and will change as the race goes forward.


by BDM on Wed May 02, 2007 at 05:39:28 PM EST
[ Parent ]

Re: Why is (none / 0)

OK, I have to step in on a comment like that. Of course you will conclude Clinton's lead in SC is double digits if the only polls from SC you look at are the ones when she is ahead by double digits, and ignore the three others taken during the same time period. If you average the five SC polls taken from April 9-30, it looks like this:

Clinton: 33.8%
Obama: 25.2%
Edwards: 18.6%

That isn't a double-digit lead, and it doesn't even include the poll showing Obama ahead. It is, instead, almost exactly a replication of her national lead. On a slightly different note, that similarity between SC polls and national polls makes me think that SC might not actually be that important of a state in which to campaign. Iowa and New Hampshire rise in importance...
by Chris Bowers on Wed May 02, 2007 at 06:43:25 PM EST
[ Parent ]

Re: Why is (none / 0)

Chris, I saw BDM's post with the "Moore Methods" poll.  The heading of his post was a carryover from a previous post, labelled "Re: New South Carolina poll."   I assumed he was talking about yet another SC poll that had just come out.  Therefore, that would have made 3 polls to come out within just the last few days, all showing double-digit leads.   Upon closer reading of his post, the Moore poll was for CA, so that takes some of the point away I was making.  


by georgep on Wed May 02, 2007 at 09:45:32 PM EST
[ Parent ]

Re: ARG Polls Iowa, New Hampshire and South Caroli (none / 0)

Since the first debate was seen by at most 1% of the voters, I doubt it had much to do with the poll shifts.

I think the field is gaining on the top three for two reasons:

1. Soft supporters of Clinton, Obama, and Edwards are learning more details about their positions, and backing away.

2. Previously undecided voters are learning that there are several other candidates that the mainstream media hadn't told them about.


by Lex on Wed May 02, 2007 at 03:47:20 PM EST

Re: ARG Polls Iowa, New Hampshire and South Caroli (3.00 / 0)

Dont over read a single poll.  Particularly one with a realitively small sample and and a huge number of candidates which breaks down the number of supporters for each candidate to a small absolute number.  The overall margin of error on a poll like this is about +/- 6. Wouldn't the margin of error for indivisual candidates be even higher?

Obama loosing 40% of his support in NH during a month were he exceeded fundraising expectations and got a lot of good press doesn't make sense.


by upper left on Wed May 02, 2007 at 04:00:02 PM EST
[ Parent ]

Re: ARG Polls Iowa, New Hampshire and South Caroli (none / 0)

Sure it does.  We all saw his less-than-Presidential debate performance.


by ChicagoDude on Wed May 02, 2007 at 04:53:48 PM EST
[ Parent ]

Re: ARG Polls Iowa, New Hampshire and South Caroli (none / 0)

Biden spent some time her a few weeks ago--I thought he was a pompous windbag whose ideas were 100% out dated but maybe not everyone did.


by MNPundit on Wed May 02, 2007 at 07:58:45 PM EST

NEW NEW JERSEY POLL (none / 0)

Clinton     40%
Obama     23%
Edwards   12%

http://strategicvision.biz/political/new jersey_poll_050307.htm


by ChicagoDude on Wed May 02, 2007 at 10:31:37 PM EST

Obamamaniacs will hate these polls (none / 0)

It will spoil their fantasies that Obama is leading by 20.


by dpANDREWS on Wed May 02, 2007 at 10:36:50 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.

Login
Username:

Password: