I agree with your hesitation at accepting the Ipsos poll because it is of all Americans instead of registered voters. However, Rasmussen's use of "likely voters" is just as suspect at this point. Likely voter screens don't work well 8 months from an election, and most pollsters won't switch to them until 3 months or so before the election. Gallup says that their likely voter model is only proven to work in the last poll before an election, and I have a hard time believing that Rasmussen's is better than Gallup's. Basically, a "likely voter" screen now is meaningless.
The best course to take right now is to throw out both "all American" polls and "Likely Voter" polls, and go with the "registered voter" model. I've done that using pollster.com's graphs, and it gives you a tied generic ballot.
As I said in my other comment, Rasmussen's Generic House Ballot results were radically out of line with those of other pollsters at the beginning of 2008, just as they are right now. In that case, Rasmussen's numbers gradually changed to match those of other pollsters; the results from Gallup (other than one freak poll that actually showed Republicans ahead) and WSJ didn't change much over that time.
It will be interesting to see if Rasmussen follows the same pattern this time. If he is showing a tied or nearly tied generic House ballot by June, then I would be inclined to guess there is a large house effect at the beginning of election years that disappears as we get closer to the election.
Rasmussen polls are pretty good within a couple of months of the election, but this far out they tend to tilt pretty heavily toward the Republicans. Back in 2008, in their January through March generic house ballot polls, Rasmussen showed Dems up by 4 or 5 at a time when Gallup and the Wall Street Journal poll was showing Democrats up by 13-15. The difference disappeared gradually in April and May and by June they were right at the average of other pollsters.
I wouldn't be at all surprised to see the pattern repeat itself this year, with Rasmussen tilting Republican significantly over the first three months of the year, then gradually coming into line with the rest of the results.
I'm as disappointed as anyone about this election. I felt physically ill Tuesday and have been walking around muttering "They have no idea what they've done!" ever since.
But it's over. We lost this time, but look at what we accomplished:
Bush was an incumbent president in war time. He once had an 80% approval rating. Millions of people formed a close, intensely emotional bond to him as a result of 9/11 (I don't know why, but they did). Millions more are still so afraid that they were terrified at the thought of any change. Conservatives have spent 40 years building a network of interest groups, PACS, and organizations to support their candidates. They have 2 "news" channels. We joined this battle less than 2 years ago.
And we fought them to a near draw.
They are at the height of their strength. We are just getting started.
They will never be this strong again. They will never again have the advantages they had this time.
We will get stronger.
They are undermining their own position already by claiming a "broad mandate" when 49% of the people voted against them. They are trampling the rights and interests of minorities. They have dedicated themselves to pursuing policies for which they have no mandate. They are setting up 2006 for us already.
We can beat them. But only if we focus on the future, instead of on the past.
It's part of being a liberal. We want to save everyone, even those who don't want to be saved.
Personally, I have moved beyond that. I will not do anything to "unite" the country. I will not move. If Bush & Co. want to destroy the country they can do it without my cooperation, without me compromising, without even civility from me.
Those people are fascist. They are destroying everything that is good about the country. I want to strengthen the rift that separates them from me, not bridge it.
i saw an announcement that Zogby and Rock the Vote were conducting a first of its kind experimental poll of cell phone users. I don't know how accurate it is, since it is a first of its kind effort, but it is a good experiment to conduct.
Zogby continues to be on the cutting edge of polling.
This was a subset of their national poll, so the margin of error has to be huge. I'm thinking probably larger than the 11% that are switching from Bush to Kerry.
Zogby's final state polls weren't too bad, except for California. But in the days leading up to the election they were all over the map. Right now, they don't mean anything.
Michigan and Pennsylvania are as safe for Kerry as any state can be. Don't worry about Zogby's tracking polls, they are not as accurate as a traditional multi-day poll.
I read the article and it certainly sounds like Zogby's comments were misinterpreted. I believe that Zogby said it is "more likely" that Bush will be re-elected now than it was in spring, not that it was more likely that Bush would be elected than that Kerry would be elected.
He said, according to Novak, that Bush has consolidated his base. Bush hadn't done that earlier in the year. But his base alone won't be enough for him to win in a high turnout year like this one is likely to be.
No. Tracking polls are pretty erratic. A similar thing happened in 2000's Zogby tracking poll in Michigan. It went from Gore +10 to Gore +1 in 3 days. Not likely.
Similarly, this result is unlikely. I fully expect it to swing back to kerry very soon.
Thanks for doing the work. I think Kerry will get more of the vote than Gore did because nader is going to get less, and of course the accidental votes for Buchanan won't make any impact at all.
Those numbers just don't make sense, no matter how you look at them. I guess the L.A. Times shouldn't even try polling on a state level. They were wildly off on the recall election, if I remember correctly, and that was their own state.
jeromearmstrong Our Polarized and Money-Driven Congress: Created Over 25 Years By Republicans (and Quickly Imitated by Democrats http://bit.ly/ewXlXI #bblue
I agree with your hesitation at accepting the Ipsos poll because it is of all Americans instead of registered voters. However, Rasmussen's use of "likely voters" is just as suspect at this point. Likely voter screens don't work well 8 months from an election, and most pollsters won't switch to them until 3 months or so before the election. Gallup says that their likely voter model is only proven to work in the last poll before an election, and I have a hard time believing that Rasmussen's is better than Gallup's. Basically, a "likely voter" screen now is meaningless.
The best course to take right now is to throw out both "all American" polls and "Likely Voter" polls, and go with the "registered voter" model. I've done that using pollster.com's graphs, and it gives you a tied generic ballot.
As I said in my other comment, Rasmussen's Generic House Ballot results were radically out of line with those of other pollsters at the beginning of 2008, just as they are right now. In that case, Rasmussen's numbers gradually changed to match those of other pollsters; the results from Gallup (other than one freak poll that actually showed Republicans ahead) and WSJ didn't change much over that time.
It will be interesting to see if Rasmussen follows the same pattern this time. If he is showing a tied or nearly tied generic House ballot by June, then I would be inclined to guess there is a large house effect at the beginning of election years that disappears as we get closer to the election.
Rasmussen polls are pretty good within a couple of months of the election, but this far out they tend to tilt pretty heavily toward the Republicans. Back in 2008, in their January through March generic house ballot polls, Rasmussen showed Dems up by 4 or 5 at a time when Gallup and the Wall Street Journal poll was showing Democrats up by 13-15. The difference disappeared gradually in April and May and by June they were right at the average of other pollsters.
I wouldn't be at all surprised to see the pattern repeat itself this year, with Rasmussen tilting Republican significantly over the first three months of the year, then gradually coming into line with the rest of the results.
But it's over. We lost this time, but look at what we accomplished:
Bush was an incumbent president in war time. He once had an 80% approval rating. Millions of people formed a close, intensely emotional bond to him as a result of 9/11 (I don't know why, but they did). Millions more are still so afraid that they were terrified at the thought of any change. Conservatives have spent 40 years building a network of interest groups, PACS, and organizations to support their candidates. They have 2 "news" channels. We joined this battle less than 2 years ago.
And we fought them to a near draw.
They are at the height of their strength. We are just getting started.
They will never be this strong again. They will never again have the advantages they had this time.
We will get stronger.
They are undermining their own position already by claiming a "broad mandate" when 49% of the people voted against them. They are trampling the rights and interests of minorities. They have dedicated themselves to pursuing policies for which they have no mandate. They are setting up 2006 for us already.
We can beat them. But only if we focus on the future, instead of on the past.
Personally, I have moved beyond that. I will not do anything to "unite" the country. I will not move. If Bush & Co. want to destroy the country they can do it without my cooperation, without me compromising, without even civility from me.
Those people are fascist. They are destroying everything that is good about the country. I want to strengthen the rift that separates them from me, not bridge it.
I hope to see you all at the barricade.
Kerry 311
Bush 226 (because the West Virginia elector will withhold his vote).
Besides, this is the same guy who was predicting an easy win for Bush in 2000, and we know what happened there.
Zogby continues to be on the cutting edge of polling.
Or something like that.
Congratulations Chris! You deserve the fame and support.
He said, according to Novak, that Bush has consolidated his base. Bush hadn't done that earlier in the year. But his base alone won't be enough for him to win in a high turnout year like this one is likely to be.
Similarly, this result is unlikely. I fully expect it to swing back to kerry very soon.