RSS 2.0 FeedLatest Gallup Poll: More Of The Same
by Doug MataconisAs the Republican candidates prepare to debate yet again tonight, Gallup has released it’s latest poll and it shows little change in the race, and little movement for the Paul campaign:
PRINCETON, NJ — As the Republican presidential candidates square off in the latest in a series of debates — this time in Dearborn, Mich. tonight — new Gallup polling shows former New York City mayor Rudy Giuliani continuing to hold onto his status as frontrunner, 12 points ahead of former Tennessee Senator Fred Thompson and 16 points ahead of John McCain. Thompson — for whom this will be the first national debate since officially declaring his candidacy — is still less well-known to Republicans nationwide than Giuliani and McCain, suggesting that the debate could be an important opportunity for him to fill in the blanks in the minds of GOP primary voters. The poll also finds former Arkansas governor Mike Huckabee moving within two points of former Massachusetts governor Mitt Romney, whose nationwide support among Republicans remains mired in single digits.
If there has been a change of note over the last several months, it has been the failure of Mitt Romney to capitalize on his mini-surge when he won the Iowa straw poll in early August. John McCain has, at the same time, recovered his standing after slipping in August. There has been a slight but steady increase in support for Mike Huckabee, despite his very low name identification. Huckabee has gone from 1% of the vote as recently as May to 7% today, within two points of the better-known (and much better financed) Mitt Romney, and clearly separating himself from the other “minor” candidates competing for the GOP nomination.
And for the Ron Paul supporters out there wondering why the boom on the Internet isn’t reflected in the polls, there seems to be a very good reason:
Ron Paul, who made waves last week with a strong fundraising performance in the third quarter, still has not gained any traction in the polls. The Texas Congressman and former medical doctor is now supported by just 2% of Republicans nationwide, even after the widespread reporting of his recent $5 million campaign haul. To date, his top performance in nationwide GOP polls has been 4% in a mid-September poll. Paul has a very substantial name identification problem among Republicans, 72% of whom say they don’t know enough about him to have either a positive or negative opinion.
What this means, folks, is that most people don’t spend all their time on blogs and cable news channels, both of which have been giving Ron Paul a lot of coverage over the past several weeks. It also means that the campaign needs to start spending some of that money it’s holding in reserve to generate public awareness. It doesn’t matter how good your ideas are if nobody has heard of you.
You also have to remember that those polls are conducted using people that have been identified as “likely Republican voters.” Which usually means you voted for George Bush in the last election etc etc. Also these phone polls are conducted using landlines. It seems like most Ron Paul supporters don’t meet alot of the requirements to be polled, and many (due to the age group) use cell phones as their main number, and would not be contacted regardless (which is my case.)
Comment by David McLeod — October 9, 2007 @ 8:00 amDavid,
Yes, these are the same criticisms that others have been noting for months, but I’ve seen no real evidence that the polls are so statisically unsound that they aren’t within the margin of error.
Comment by Doug Mataconis — October 9, 2007 @ 8:28 amYou’re right, and they started to do just that a few days ago. But they’re not going to do it in 50 states, so the national polls won’t be particularly relevant for Paul until the end of the year.
The key is watch the effectiveness of his ads in states like NH, IA, and MI.
Comment by Jeff Molby — October 9, 2007 @ 9:27 amI agree with Doug that Ron’s biggest problem is likely his name recognition. The majority of people do not watch cable news networks or go to the internet this early for political information. That is the next big hurdle for the Paul campaign is getting his name out and educating the voter about how he stands on the issues. In the next few months people will start trying to find out a little bit of information about candidates and the Paul campaign needs to be there providing it.
Saying that I also believe that David is correct as well. These polls are obviously skewed to the old guard in the party. In the past they may have done well in determining the correct % of vote but things are changing. As David noted these are done normally by land lines and many younger people either do not have land lines or if they do have caller ID and don’t answer numbers they do not know. In addition these polls are not picking up the new people to the party that are in large part Paul voters. In open primaries Paul may also be able to bring over more outside votes than is typical in a republican primary.
My guess is that he is likely more in the 4-8% range at the current time with that number likely to go up with more name and issue recognition.
Comment by TerryP — October 9, 2007 @ 9:37 am[quote]Paul has a very substantial name identification problem among Republicans, 72% of whom say they don’t know enough about him to have either a positive or negative opinion.[/quote]
I agree! Dr. Paul, spend our money! We don’t have time for another “surprising upset” in fundraising. You got the big boost, now use it fully to your advantage!
Comment by Blowback — October 9, 2007 @ 9:39 amDoug,
If you haven’t seen the evidence of statistical unsoundness, then you must have had your eyes (and ears) closed.
The Plain Old Telephone Surveys (POTS) have had a terrible track record in relation to the straw polls. I am not even talking about Ron Paul’s results, and I am not talking about straw polls which are open to every state fair attendee.
I am also not talking about cellphone bias, or the fact that Ron Paul’s name is routinely left off of POTS (and then those POTS are averaged with other POTS which artificially depress Ron Paul’s Real Clear Politics poll averages)
In IA, TX, and MI, the polls were limited to GOP members. In TX, it was very limited – not just GOP members, but past GOP state or national delegates. In theory, these straw polls should have had results within the MoE of the POTS.
Let’s see what really happened.
IA – ABC POTS had McCain and Huckabee even at 8% among likely GOP straw poll attendees the week prior to the straw poll. McCain got 1%; Huckabee got 18%. Both results were far outside the margin of error (MoE) for that POTS (4.5%). Romney did well, within the MoE of his POTS numbers, but Giuliani did terribly (well below Tommy Thompson)
TX – Duncan Hunter won the straw poll with 41%. He had been polling in the low single digits (even lower than Ron Paul) in POTS. Again, Giuliani did very poorly in the straw poll with 6%, no where near his commanding 26% in the POTS. Theorhetically, this straw poll should have been closest to the POTS numbers due to the restrictive straw poll criteria. If anything it was the least conforming.
MI – The Mackinac straw poll had the order:Romney 39%, McCain 27%, Paul 11% (106), Giulaini 11% (104), F. Thompson 7%, Huckabee 3%. But the Real Clear Politics POTS average had this order: Romney 26.7%, Giuliani 18.3%, Thompson 14.0%, McCain 13.0%, Huckabee 4.3%, Ron Paul 1%. Of course Romney does well in MI, his father used to be governor. but how did McCain do so well in the straw poll, while doing so poorly in the POTS?
Here then, are 3 straw polls whose attendees were as likely to be GOP primary voters, as any POTS respondants.
Not one of the POTS results is anywhere on the same planet as the straw poll.
Conclusion? The POTS bear little or no relation to actual voting patterns this far from the primary. Plain Old Telephone Surveys are statistically unsound, open your eyes.
I’m not saying Ron Paul is going to win based on this. Ron Paul’s numbers are in the 11-16% for these straw polls. I am saying that Ron Paul’s support level is WAAaay above the 1-4% that most POTS place him at.
Later.
Comment by Kevin Houston — October 9, 2007 @ 9:54 amHistorically, “POTS” are a lot more accurate than straw polls. Straw polls count people who care enough to attend a fairly meaningless event while phone surveys measure people who are likely to vote on primary day.
Comment by Bob — October 9, 2007 @ 10:03 amUh Bob,
I think you got that backwards.
Kerry in Dec 2003 was at 4%, then he won Iowa Caucus and suddenly went to 53% in national POTS.
POTS measure name recognition and “preference”, they don’t measure who is motivated enough to get up off the couch and go to the polls.
I agree with you when the straw poll is at a state fair, or some other walk-by event (people were already up off of their couches in such cases) but the three specific state straw polls I mentioned required you to be registered GOP, and pay money, and show up (often at a location physically distant from one’s home – such as the Mackinac straw poll for example.) The people going to such events have spent time researching the candidates, while POTS respondants were thinking about something else entirely when the pollster called them.
In short, straw polls are far more like voting on primary day than answering your telephone and saying the first thing that pops into your head.
Comment by Kevin Houston — October 9, 2007 @ 10:30 amRon Paul has made substantial progress in the polls; it’s just been slow. He started in the 0-1% range in March, and now hits 3-4% nationally, and 4-6% in Iowa and New Hampshire.
That doesn’t look like much of course, but it was done the hard way, with very little TV exposure. It’s unrealistic to except his numbers to shoot up in the first week after he got any kind of positive press at all. People will take longer to make up their mind about him.
I agree that he needs to hit the airwaves, and as I understand it, that is the plan. If he’s already at 5% in New Hampshire, and can still triple his name recognition, he becomes a major threat. Why? Because his supporters are much more enthusiastic than those of the other candidates, and will turn out at a much higher rate.
A lot of commentators ask if Ron Paul’s Internet support will translate into votes, but the question they should be asking is whether Rudy Giuliani’s support in random telephone surveys will translate into votes. Ron Paul has far more actual supporters than any of the other candidates.
Comment by Craig — October 9, 2007 @ 10:55 amVery good point, Craig.
I just can’t understand how a thrice-divorced, gun-grabbing, cross-dressing, pro-public-funding for abortions, mayor of NYC can possible be getting 30% in national POTS among GOP faithful.
This is why I think the straw polls are more indicative than the POTS – the straw poll participants have actually looked at Rudy’s record. With Dobson and Chuck Baldwin saying they will vote 3rd party if he is nominated, his “electibility” card is trashed. Giuliani is the one candidate who can’t win, because 1/2 the GOP base won’t support him on a scare dare.
However, as encouraging as some of the latest state POTS results are for Ron Paul, POTS still don’t mean a thing this far from the first primary.
Here’s Edwards, talking about the same thing on the Democrat side. We all remember the exciting Bush/Dean election from 2004, Right?
http://apnews.excite.com/article/20071009/D8S5E5601.html
Later.
Comment by Kevin Houston — October 9, 2007 @ 11:17 amIn 1999 Keyes won the Alabama straw poll. Paul won it this time. Straw polls are largely meaningless. Most people realize how meaningless straw polls are and don’t bother with them. But the same people who don’t turn out for a straw poll will turn out on primary day.
Comment by Bob — October 9, 2007 @ 12:08 pmI do agree that phone polls only measure what people think at the moment. But right now Paul’s numbers aren’t encouraging. The last Iowa poll had Paul at 4% with 71% saying that they had ruled him out. Nationally his favorable rating is 14% his unfavorable rating is also 14%. Every other candidate has a much higher favorable to unfavorable rating.
I just can’t understand how a thrice-divorced, gun-grabbing, cross-dressing, pro-public-funding for abortions, mayor of NYC can possible be getting 30% in national POTS among GOP faithful.
You can’t? He was about the only public figure who didn’t look like a bedwetter on America’s worst day in a century or so. That counts for a lot.
Comment by Joshua Holmes — October 9, 2007 @ 12:08 pmKevin,
Comment by Bob — October 9, 2007 @ 12:27 pmI’ll concede that I don’t think our fellow Republicans will nominate a pro-abortion, anti-gun, weak on illegal immigration candidate like Rudy. I think Romney or Thompson is more likely to get the nomination.
Bob,
I see you totally side-stepped the straw polls I was refering to. I wasn’t talking about the Alabama straw poll, I was talking about TX, MI, and IA.
Joshua,
*looking* good, and *doing* good are two totally different things. Yes, Rudy Giuliani looked great on 9/11. But he didn’t really do great. Talk to the Firemen, Police and Paramedics. Talk to the people with 9/11 related lung problems. Ask why the emergency command center was placed in the same building that was targeting in 1993. and on and on and on..
Rudy’s support may be wide, but it is as shallow as a puddle on the flat prairie. Once people look past the images, they will see things they don’t like. The religious conservatives are already revolting over revolting rudy.
Comment by Kevin Houston — October 9, 2007 @ 12:29 pmFor the purpose of this comment, assume that POTS are flawed because they only poll “likely Republican voters” over landlines. They fail to account for the substantial part of the populace that uses a cell phone exclusively or does not publicly identify as a “likely Republican voter”.
Scientifically speaking, how could an inexpensive, nationally scaled poll be designed to accurately reflect the sentiments of the people who actually will vote on primary day?
I know of none, but I’m an engineer and polling methodology is certainly not my forte.
Comment by Brian T. Traylor — October 9, 2007 @ 1:12 pmOne thing that troubles me about Gallup/Rasmussen/AP polling is this:
As libertarians, we often talk about market mechanisms that work to benefit consumers, and the intangible value of reputation for a firm is paramount for the company that wishes for its product to be purchased or highly regarded in the marketplace.
That said, were POTS to be wholly inaccurate, why would a polling company with a household name (and reputation to protect) like Gallup risk said reputation putting out a poll using flawed methodology?
Personally, I’m inclined to think that the polls are inaccurate, but I’m aware that I suffer from selection bias.
Comment by Brian T. Traylor — October 9, 2007 @ 1:27 pmI think the best way to do that would be to do weekly polling of the same (extremely large) group. Then this group would come to expect the poll, and will have given serious thought to their answers prior to being called. Since they know it’s just one company, even cell-phone users can be included or opted out as they wished. perhaps people could sign up over the internet as well.
The first few rounds would be pretty much meaningless, but as it went on, the answers would get closer and closer to reality. The pollsters could have access to the demographic data to adjust for those that refuse to be bothered, and the respondents could become more trusting about who is doing the poll and what their true motives are.
A few hundred (thousand?) IVR systems, and a large database of registered voters (cross-ref’d with list of who actually voted last time.)
Then Mrs. Jones would come to expect her Tuesday afternoon polling call, while Mr. Smith might get his on Wed. evening. But each respondent would get 1 call every week, at about the same time.
The larger the sample, the more it will reflect reality. POTS are small because they used to have to do them by hand. One could throw in a few American Idol / Fantasy Football questions, for the respondents, and a few marketing questions for the advertisers (to keep it interesting, and help pay for the equipment and analysis of the data.)
Another advantage would be that the polling company could track shifts in attitude on a per respondent basis. If a bunch of long-time Democrats in one particular area, or demographic start voting for the weakest Republican, then one can speculate that they are just mischief makers.
But if lots of different people from all over the map (geo, or demo)graphically start voting for the weakest Republican, then perhaps he’s not really the weakest Republican.
Later.
Comment by Kevin Houston — October 9, 2007 @ 1:29 pmBrian asks:
Because the economic incentive is on the polling company getting the guy with the most money to pay for the poll. In a market with more than one polling company, Giuliani is going to commission his internal polling from the company that is giving him the best results.
If they don’t think Ron Paul has any money, then they won’t masssage the data to benefit him.
They don’t have to be right this far out. They only have to be close the week prior to election day. There are many ways to do this, by changing the tightness of the screen, or by polling different lists (as some have been accused)
Later
Comment by Kevin Houston — October 9, 2007 @ 1:35 pm*looking* good, and *doing* good are two totally different things.
It’s politics. Perception is reality. Almost nobody knows anything about the NYFD or NYPD complaints about Giuliani. I don’t, and I’m far more politically savvy than your average primary voter. Hence, the image of Giuliani acting like an adult on America’s worst day sticks in the mind of the average voter.
Rudy’s support may be wide, but it is as shallow as a puddle on the flat prairie. Once people look past the images, they will see things they don’t like.
Since people pay no penalty for “voting wrong”, they have little incentive to look past image to substance.
Comment by Joshua Holmes — October 9, 2007 @ 1:41 pmPhone polls are more like voting on primary day than straw polls because phone polls make an attempt to reach a cross section of likely voters while only a small section of those voters care about who wins a straw poll. In Iowa Rudy, McCain and Thompson didn’t compete even with those 3 out of the race Paul finished 5th. In Paul’s home state (Texas) with only Hunter competing Paul finished 3rd (Thompson who wasn’t a candidate also beat Paul). I haven’t studied the Michigan results and therefore won’t comment on that straw poll.
Comment by Bob — October 9, 2007 @ 1:48 pmBob,
That may be true of general elections, but I don’t agreed that’s the case with primary voters.
Primaries attract a much smaller participation rate than general election. And because the primary is held on different days in different states, there is no nationwide GOTV prod from the national pundits like there is during the general election. I think the primary voter demographic will look a lot like the straw poll demographic.
This year, the primaries will have even fewer of the usual GOP base, because for the most part, they are really disgusted with the rampant over-spending and just plain tired of the war as well.
The time is ripe for a general take-over of the GOP, and Ron Paul is just the guy to do it.
Later.
Comment by Kevin Houston — October 9, 2007 @ 2:14 pmKevin,
I think your point that “they don’t have to be right” wins me over. At this point (still 3 months out for the earliest primaries), there are no penalties for being wrong.
Comment by Brian T. Traylor — October 9, 2007 @ 2:23 pm[...] Posts on Gallup Poll results can be found here (10/9), here (9/18), here (9/10), and here (8/20) [...]
Pingback by Below The Beltway » Blog Archive » The Latest Gallup Poll: The Song Remains The Same — October 17, 2007 @ 11:06 am