Topic: Presidential Campaign 2008
Ron Paul Moves Up Six Points in Iowa Poll by American Research Group The latest scientific poll from American Research Group shows that Ron Paul's support in Iowa over the past four days has more than doubled from 4% to 10%. Is this the beginning of Paul's sprint to the finish line?by Walt Thiessen
(Libertarian)
Wednesday, December 26, 2007
One of the biggest criticisms Ron Paul has received has been his apparent inability to penetrate into double digits in scientific polls. Many reasons have been given for this both for and against the Paul campaign. However, polling results just released by American Research Group Inc suggest that the scientific polls may finally be catching up to where Paul's supporters claim he has been all along.
Just four days ago, a poll ARG took Dec. 16-19 showed Dr. Paul with the support of just 4% of Iowa caucusers. Four days later, a poll taken Dec. 20-23 by this same polling organization shows him at 10%, more than twice his previous support level. The poll shows nearly all of Paul's main opponents dropping in their support, including front runner Mike Huckabee who lost five points to 23%, down from 28% four days prior. The only main opponent who rose was Mitt Romney, whose support rose four points after an 11 point drop from their prior poll. This is most likely a bounce after a fall due to the controversy regarding Romney's religious beliefs.
Even more interesting, the same poll shows that Dr. Paul's support among women in Iowa is slightly higher than men, as 11% of women are expressing their support for Paul in the poll.
The poll doesn't say what its statistical margin for error is. Similar polls typically have about a 4 point error margin. If this poll has a similar sample size, it means that Paul's support has increased by more than the margin for error, which likely indicates that it is at least partly a genuine move upward and not merely a statistical anomaly.
What is even more interesting is that ARG typically reports that about 80-90% of the support for the top candidates is not likely to change. Yet, the percentages of support for each candidate appears to be swinging dramatically. This suggests that the "strong" support may not be as strong as the pollster thinks it is.
ARG also conducted a poll on Dec. 20 in which they found that 45% of Republicans are now dissatisfied with George Bush's economic leadership, up from 25% in November. That's a dramatic increase. Given the fact that only Ron Paul is discussing economic issues directly and in detail, this could be part of the reason why his support has jumped so much in just four days.
It's dangerous to read too much into these numbers, given the volatility that the scientific polling numbers have been showing from all pollsters over the past few weeks and months. Still Paulites and the Paul campaign have to feel encouraged, not the least because with this poll Paul appears to have finally broken into double-digits. With the Iowa Caucuses just eight days away, this could be the beginning of a Paul wind sprint to the finish line in his attempt to catch the frontrunners. Not bad for a 72-year-old ob/gyn.
Did you like this article? If you did, Thumb It! 267 thumbs so far
2007 Walt Thiessen, all rights reserved.
Published: Wednesday, December 26, 2007
Last modified: Wednesday, December 26, 2007
The views expressed in this
article are those of Walt Thiessen only and do not represent
the views of Nolan Chart, LLC or its affiliates. Walt Thiessen is
solely responsible for the contents of this article and is not an
employee or otherwise affiliated with Nolan Chart, LLC in his/her role as a columnist.
The margin of error applies to both polls you cite. If the earlier poll was lower than actual support by 4% and the latest poll was higher than actual support by 4%, that leaves a range of 6 - 8% where the two polls could be in agreement, measuring no change at all or even reduced support for Ron Paul, say from 8% to 6%. Still, I am happy to see the possibilty of upward movement and am anxious to see the next few survey results.
So, Ron Paul's numbers are up when polling people who are likely-caucus goers that turned out in 04 and 06 when turnout was very low? This is even better than I thought. If he can convert those people, then add in all the new caucus goers and he might take second or even first. I don't want to get my hopes up, but how many of his supporters are unlikely to show up, one or two from the odd car accident? Someone mentioned 70% would be great turnout because the others might get 20-30%. I'd be shocked if only 70% of Ron Paul supporters actually showed up. Shocked. I'd say 99% is more like it.
Jim: Margins of error are more of a proportional thing. Generally a poll result 50% with a margin of error of +-5 means anywhere between 46-54%
a poll of 5% doesn't mean 0%-10% support thats rediculus assuption, more like 4.6-5.4% though I would guess its not a direct linear relationship and it could easily go as far as 3-7%
Even just 10% of people who voted for Bush in the 04 primaries is a phenomenal number! I figured we'd have to be in the middle of another Great Depression before something like that happened. Maybe this is proof that we are in a depression and not a recession??
Posted By: Walt Thiessen
Date: 2007-12-26 11:52:36
Jim and Nick are technically correct, but they are ignoring the volatility in the numbers. That's what makes the new numbers so intriguing. Margin of error doesn't play a role when the numbers are scarcely changing. When they change as regularly and as dramatically as they have been, it means that all support is extremely soft.
I didn't claim that margin for error doesn't span 4 points either side of the figures cited. I claimed that when the changes exceeds the numeric margin for error, it means that you can't just dismiss it as being a statistical anomoly. I agree that the final proof is when a move exceeds 8 points with a 4 point margin, but I disagree that a 6 point move within a 4 point margin is no more than a statistical anomaly, particularly when all the numbers keep changing so fluidly. When numbers are as soft as these are, there is plenty of room for change to take place, and a shift of this kind can portend an actual trend. That's what I am claiming, and no more.
I like these numbers too. Based on my meet up group of over 100 people we have about 10% who might have voted for Bush in 04 and that have land line phones. If this is any indcation of what is going on at the grass roots level. That 10% is consistant with ou rlocal sample of supporters. People have been saying for months that if he can get into the double digits with people who tow the party line then he will take the prize with ease with his surge of support from new voters, Libertarians, Contsitution Party and Democrats who realize their party .
So the 90% of the rest of us that were Libertarians, Democrats, and non voters could flood the primary with new blood.
Another thing to remember is that everyone who supports Ron Paul is not in a meetup group, not sending him money, and not on my space or any of the rest. So it really is bigger than even the internet is showing. My mom gives a Ron Paul slim jim to everyone she meets and has never even toughed a computer.....
Based on my 90% 10% view Dr. Paul should win by a landslide being conservative about it. I doubt that the others are building the party like he is so their numbers are probably good in relation to th at voting block from 4 years ago. Look at how Ron Paul's fundraising snowballed and now is compounding. ( 2nd Q doubled 1stQ, 3rdQ about doubled 2nd, 4thQ will be 4x s )
I was looking at the polls for the Kerry election in 2003-4 on CNN, if you care to look. In December Kerry had 13% and Dean about 46. By Jan he was tied with Dean and Clark went down to single digits from about 22. The next poll in January and Dean was at 8, Kerry at about 46, something like that, I forget. I would think a simple street poll would do better. Do people actually change their mind that fast? Theoretically the meetup groups should have numbers when they canvas for their areas. Sometime they match well and sometime this early not. It depends I suspect.
i have read an email about the Israeli Defense Firm to tally the Iowa Caucus. i need to know if anyone else has read or can verify reputation of the site.
Want to comment on this
article? Leave your comment here. Your email address is
required to track your comment. However, we will neither
publish your email address nor distribute it to other
organizations or persons. The only reason we might use
it would be if we needed to contact you regarding your
comment. All comments are subject to our
terms of use policy.
2007 Walt Thiessen, all rights reserved.