The Candidates
Here is the latest batch of polling from ARG:
American Research Group Iowa Poll, conducted April 27th-30th, 2007
- John McCain 26%
- Rudy Giuliani 19%
- Mitt Romney 14%
- Fred Thompson 13%
- Newt Gingrich 8%
- Mike Huckabee 2%
- All others less than 2%
- Undecided 12%
And for SC…
American Research Group South Carolina Poll, conducted April 27th-30th, 2007
- John McCain 36%
- Rudy Giuliani 23%
- Fred Thompson 10%
- Newt Gingrich 6%
- Mitt Romney 6%
- Mike Huckabee 2%
- All others less than 2%
- Undecided 12%
And finally NH…
American Research Group New Hampshire Poll, conducted April 27th-30th, 2007
- John McCain 29%
- Mitt Romney 24%
- Rudy Giuliani 17%
- Fred Thompson 7%
- Newt Gingrich 4%
- All others less than 2%
- Undecided 14%
Click here for the Dems.
May 1st, 2007 at 2:56 pm
Gingrich has really dropped. Like I have said before, I think Fred Thompson jumping in keeps Newt out, or at the very least, cuts his support in half.
May 1st, 2007 at 2:59 pm
Interesting… still a long ways to go before any votes actually count.
Looks like Romney’s support varies greatly. Looks hopeful in NH, not so
great in SC, Mediocre in IA.
May 1st, 2007 at 3:56 pm
Yeah, Romney and all his money is not pulling through… Just shows that you need more than just good hair and a lot of money to win in the polls. Have you guys ever thought about how Romney reflects a lot of the doctrine that is found in The Prince, by Machiavelli?
May 1st, 2007 at 4:01 pm
I should say these three polls look *very* good for Senator McCain…
May 1st, 2007 at 4:04 pm
These polls are great news for McCain and Romney and horrible for Giuliani. The trends:
In Iowa, Romney is up 4, McCain is down 3, and Rudy is down a whopping 10 points.
In New Hampshire, Romney is up 7, McCain is up 6, and Rudy is down 2.
In South Carolina, Romney is up 1, McCain is up 1, and Rudy is down 6.
All total, Romney gains 12 points (up in every poll), McCain gains 4 (up in 2, down in 1), and Rudy loses 18 points (down in all 3 polls).
The long predicted bleeding of support from Giuliani has begun. Meanwhile, Romney is coming on strong to pick up the pieces.
May 1st, 2007 at 4:08 pm
Here’s the thing about ARG polls… They have never shown the McCain slide that all other polls do. The last Florida poll they did showed McCain with a 10 point lead on Rudy when everything single other polling firm shows the exact opposite to be true. McCain over-performs in every ARG poll that has ever been released in comparison to other polling outfits.
In fact, until after the midterms they had McCain up by 30 points in every state they polled. They were so out of whack that I would not allow them to be published here (along with Zogby Interactive polls).
Now I wish I had published them since they do not publish past polls. So you all have to take my word for it.
Just putting this out there. So take it for whatever you think it’s worth. But ask yourself why ARG’s results are so radically different from Rasmussen, Strategic Vision, Quinnipiac, SurveyUSA, Marist, Fox News/Opinion Dynamics, Gallup, etc…
May 1st, 2007 at 4:15 pm
That’s two straight polls with Giuliani in 3rd place in NH.
http://www.realclearpolitics.com/epolls/2008/president/nh/new_hampshire_republican_primary-193.html
If he can’t do better than that here in a northeast state with a history of electing pro-choice Republicans then he’s in trouble.
May 1st, 2007 at 4:25 pm
[...] http://race42008.com/2007/05/01/poll-alert-american-research-group-ia-sc-nh/ [...]
May 1st, 2007 at 4:36 pm
“That’s two straight polls with Giuliani in 3rd place in NH.”
Notice that those two polls are ARG and Zogby!
Here’s ARG & Zobgby’s results:
Now here are the other firms:
My point is that, as usual, ARG and Zogby are showing numbers that all the other firms don’t (and that is true for other states as well like SC).
That doesn’t mean that ARG and Zogby are wrong (so please read that sentence again before you savage me). It’s just means that we will all have to chose what polling firms we believe.
May 1st, 2007 at 5:04 pm
Hello-
Just some observations:
I appreciate the fact that Kavon called attention to the fact that Senator McCain seems to do so much better in ARG’s polls than he does in others. I’m trying to figure out why that’s the case:
If you go to my new-media site - just click “Campaignia.org Publisher” to do so - I have each poll in a separate post. (FYI: Race42008.com was cited as the source for “calling it to Campaignia’s attention.”) I also compared the April polls to the same ARG polls in previous months.
In Iowa, McCain and Rudy were in a dead heat in March, and now they have McCain up by 7 (statistically significant).
In NH, McCain and Romney remained in a dead heat, as they had in March. Both showed McCain with a small lead (5% in April, 6% in March), but according to their own calculations, those are dead heats.
In SC, McCain now has a 13% lead over Rudy (obviously, very statistically significant), and he had a 6% lead in February (but once more, that was too narrow to be anything other than a dead heat.
So here is the big question that I’d like to put forth to the group (I can’t figure out the answer, so anyone else’s theories are welcome):
Why does ARG show McCain in so much of a better light than every other pollster? The weight of the evidence, obviously, is that all of the other polls are right and ARG is wrong.
The reason why I view this as a conundrum, is that it’s hard for me to see how a genuine, reputable pollster (assuming, of course, that the poll has no overt pro-McCain bias) could consistently overestimate support for ONE candidate out of a multi-candidate field. Assuming that the samples are genuinely random, there’s no reason to think that McCain partisans would be overrepresented. It would be different, let’s say, if his support is concentrated in cities and they were overrepresented, but there’s nothing to my knowledge that his supporters are concentrated in any one area within those states.
If anyone can shed some light on this, that would be great.
Sincerely,
Publisher, Campaignia.org
May 1st, 2007 at 5:08 pm
Kavon, that’s a good point. I didn’t see your post about ARG and Zogby before I posted that.
Real Clear Politics currently lists 12 NH polls, 6 from ARG and Zogby and 6 from other firms. Here are the average results.
ARG/Zogby
McCain 26.5
Giuliani 20
Romney 18
Others
Giuliani 30
McCain 28.3
Romney 16.7
So there is a definite disconnect on Giuliani.
It’s also worth noting that 3 of the ARG/Zogby polls include Fred Thompson while only 1 of the others does. The ARG/Zogby polls are also more recent on average. Those are also factors that could contribute to the differences.
I don’t have access to historical results to compare the records of the pollsters. It would sure be interesting to see how they did in the 2000 NH primary compared to the actual results.
May 1st, 2007 at 5:18 pm
Publisher - Campaignia.org,
I have been researching (or I should say trying to research) that question for over 10 months now. I have been able to find very, very little about ARG’s methods.
I became interested when they started releasing their very first GOP polls that showed McCain at or approaching 40% in multiple states with every other candidate in single digits (including Rudy).
Let us know what you find…
May 1st, 2007 at 5:27 pm
Thanks for posting these poll results Kavon, I’ll be very interested to see how the polls read a week or so after the first primary debates.
May 1st, 2007 at 5:29 pm
American Research Group vs. Rudy
http://www.realclearpolitics.com/printpage/?url=http://www.realclearpolitics.com/blog/2006/06/american_research_group_vs_rud_1.html
May 1st, 2007 at 5:29 pm
Exposed: The American Research Group’s Anti-Giuliani, Pro-McCain Bias
http://giulianiblog.blogspot.com/2006/06/exposed-american-research-groups-anti.html
May 1st, 2007 at 5:42 pm
Argo,
I had totally forgot about RB’s post on this exact topic (I am “The Bij” if anyone hasn’t figured that out by now), thanks for posting that.
You can see how far back my suspicion of ARG goes.
May 1st, 2007 at 6:48 pm
So is the race between just the top 4 now? Can we pretty much write off Gingrich (Thompson surely takes out the “blast from the nicer past” vote), Brownback, Huckabee et al?
Most of the others were aiming for a big blow in Iowa to launch their run to the WH. But polling at 1-2% now does not bode well at all for any of the small players.
Just a long term thing, I remember when Romney was polling at 3-6% and people were arguing his money hadn’t even moved him out of single digits. Well, he’s now up to 15%+, trending in the right direction (though still below Rudy & McCain). I think long term he still looks a good bet…
May 1st, 2007 at 6:51 pm
In the race for the Dems Biden seems to have climbed (admittedly not massively) from 1-2% to 6%. Is that because of his debate performance, or was this poll taken before that?
May 1st, 2007 at 7:11 pm
Although this poll isn’t reliable (which isn’t saying much) it does seem to have caught the general movement of the race (McCain steady, Thompson surging, Rudy leveling off his peak and Romney all-over-the-map).
I wonder if Thompson missing the first few debates will actually be an asset if they turn ugly. At this point the GOP is really having an identity crisis. It remains to be seen if there is any candidate who can unite the party. If we end up having to choose which wings of the party we’re going to keep we may as well prepare to lose.
On Thursday I’m looking for someone that can unify the party by appealing to all of them on their important issues while presenting a compelling vision of what the new Republicanism is going to be.
May 1st, 2007 at 7:36 pm
Depends if any R’s take a page out of Gravel’s playbook? It would be naive to think they wouldn’t, especially if Gilmore, Huckabee, T. Tancredo, the Dr. and T. Thompson are there. They have nothing to lose, so why wouldn’t they take a chance.
May 1st, 2007 at 7:55 pm
Isn’t ARG the outfit that proclaimed a month or two ago that McCain was “tanking”?
May 1st, 2007 at 8:13 pm
Hello, all-
I think I may have part of the South Carolina discrepancy explained, although that’s just one piece of the puzzle. It appears that the inclusion of Fred Thompson in the April poll (he hadn’t been included in either December 2006 or February 2007), is what changed.
According to ARG’s own site, they have their last three R polls in SC charted. They took place in December 2006, February 2007, and April 2007.
Among the Big Three, their support in SC has, for the most part, remained unchanged for five months.
McCain 35% 35% 36% now.
Rudy 28% 29% 23% now.
Romney 5% 5% 6% now.
Undecided - likewise - 16% 18% 12% now.
The only real movement was Thompson. After being excluded from the first two polls, his support jumped up to 10% (higher than Romney’s, and he was just there!)
Per methodology - they claim that the names are rotated, to avoid bias.
Also, I don’t think that Thompson is taking support from Giuliani. (I doubt that they’d be appealing to the same voters anyhow.) Rudy’s meandering in the 20s is (IMHO) the standard pitch-and-roll of the margin-of-error of +/- 4%.
Anyhow, food for thought. Once I find more, I’ll let everyone know.
Reactions welcome.
May 1st, 2007 at 8:45 pm
In 2000 there was only one pollster that showed the McCain’s movement in New Hampshire and that was the ARG poll. I don’t know why but I do remember it being off from the other polls but on primary day in New Hampshire when McCain won by almost 20 points the ARG poll was the most accurate.
May 1st, 2007 at 8:47 pm
Make sure you look at who is being surveyed in all of these polls. ARG generally polls
Republicans and Independents in their GOP polls. McCain has done well with both groups,
however, one could argue that Independents provide him with an added bump that the
other candidates don’t get. If you’re looking for the “purest” polling numbers look
for likely Republican primary voters and likely Independent primary voters in states
where Independents can vote in the Republican primaries.
May 1st, 2007 at 9:53 pm
“Depends if any R’s take a page out of Gravel’s playbook?”
I don’t think Gravel did himself any favours in the Dems race. as The Fix points out, voters don’t like angry candidates, even if they are angry themselves…
May 2nd, 2007 at 3:27 am
A possible source of difference between the polls is that ARG screens people out after asking the primary question. That is they ask how likely you are to vote after they’ve asked who you’re voting for.
This brings up a dilemma. Who is framing the question incorrectly the firms that cue people in to their likelihood to vote then ask who they’re supporting or the firms that ask who likely you are to vote after giving you names of candidates.
My gut suggests that offering names first is going to cause fewer voters to drop out.
May 2nd, 2007 at 5:09 am
I hope I am not the only one who think it is way to early for debates..Geez it is May for crying out loud but of 2007. People are concerned with finishing school not who is going to be President in 2008 right now. We should wait till at least October.
May 2nd, 2007 at 9:10 am
JL, the vast, vast majority of the voting population is not still in school. Sorry to burst your bubble. This long vetting process is exactly what we need. The next President will serve at least 4 years and as many as 8, so what’s one year of campaigning in comparison? We might have been wiser about GWB had he been subject to the same rigors.
May 2nd, 2007 at 9:36 am
JF,
Agreed. I want candidates to do as much campaigning as they can. I’d like them to adopt Gingrich’s suggestion, to have weekly 90 minute “dialogues” in the general election. There’s something to be said for informed choice. Maybe then we wouldn’t end up with lightweight phenoms like Obama, and inept president’s like Bush.