The Candidates
Featured Archives
Patrick Ruffini?did it with the latest Newsweek poll.? Now it’s time for Yours Truly to re-weight the latest Diageo/Hotline ‘08 poll in order to reach a more accurate assessment of the race for the White House.? Here’s the poll as it was released, with a 40D/34R/26I split:
WH ‘08 General Election Matchup
????????????? All? Dem?GOP? Ind?? 11/15????????????????????? All?? Dem?GOP?? Ind
McCain?? 48% 17% 88% 50%? 45%?????? Giuliani???? 48% 16% 87% 51%
Clinton?? 39??? 71???? 9???? 29???? 40?????????Clinton????? 40????74???? 8???? 26
Undec??? 13??? 12???? 3???? 21???? 15???????? Undec???????12????10?????5???? 23
?
McCain?? 43% 19% 77% 38% 41%??????? Giuliani???? 41% 17% 77% 35%
Obama???38??? 64??? 12??? 31??? 36?????????? Obama?????38??? 67??? 10??? 29
Undec??? 19??? 17??? 11??? 31??? 23?????????? Undec???????21?? 16????13??? 36
?
McCain?? 45% 19% 82% 39% 49%??????? Edwards????? 41% 73% 13% 30%
Edwards? 35?? 61???10??? 24??? 36??????????? Giuliani??????? 40???? 9???? 76??? 43
Undec??? 20??? 20??? 8???? 36??? 15??????????? Undec???????? 19??? 18???? 11??? 27
And here are the results when adjusted to fit the 38D/36R/26I party identification of the most recent national election, held less than three months ago:
?????????????? Edwards????????? Hillary!????????? Obamamania?
Rudy????????42R-40D????????51R-38D?????????? 43R-37D
McCain???? 47R-33D???????? 51R-38D?????????? 45R-37D
In both the adjusted Newsweek poll, and in this poll, GOP poll leaders Rudy and McCain beat the Democratic Big 3 of Edwards, Hillary, and Obama.? The Hotline poll shows McCain running a bit stronger against Edwards and Obama than Giuliani, while the Newsweek poll shows the reverse.
January 21st, 2007 at 3:03 pm
“In 20044 it was liberals who kept claiming that the polls overweighed Republicans.”
By 20,044, Cockroaches will have inherited the earth.
“It’s not a good idea to reweight samples the way you do. The number of self-declared republicans and democrats varies a lot over time.”
In 2004, turnout was 37R/37D. In 2006, it was 38D/36R. Given that those two elections were as close to polar opposites as you can get in terms of outcome, momentum, which party had the energy, etc, I think this shows that, at least for now, both parties have figured out how to GOTV to the extent that just over a third of the electorate is going to be comprised of Democrats, just over a third will be Republicans, with independents making up the difference.
I fail to see how a 2-point Dem advantage just three months ago could’ve turned into a 6 or 10-pt one overnight, especially without any intervening cause.
January 21st, 2007 at 3:50 pm
Well, I for one welcome our cockroach overlords.
Typo aside, however, I think GT is right. In immediate post-election polls, the winning party sees a bounce in the post-election polls. The reason is probably related to the fact that typical respondents claim to have voted in the past election even if they didn’t. And, since they didn’t vote (this part is my theory), they have to make up a quick lie and simply spout off the name of the candidate or party that won.
It works in mid-term elections too. Presidential polls in 1983 and 1987 all favored the Democratic candidate, despite the fact Republicans won in 49 and 40 state landslides that year. And in early 1995 Dole was beating Clinton handily at this time in the cycle.
And even more recently, early 2005 polling had Rudy and even McCain beating Hillary — in New York! So, relax: the poll is probably an accurate reflection of the public sentiment at this point in time but if it overstates Dem support as much as early ‘05 polls overstated Rep support, Rudy and McCain are well positioned to win in ‘08.
January 21st, 2007 at 3:54 pm
I think people are missing the point here. There was an election 2 months ago. The electoral breakdown was 38D/36R. The libs at Newsweek had to give the Dems an 11 point advantage in party id to get the Dem candidates to win. This is a clear case of the media cooking the books to get the results they want.
January 21st, 2007 at 4:52 pm
The worrying portion of these polls in my opinion are Obama’s numbers. As I’ve suspected for awhile, his “feel good” rhetoric is resulting in a significant amount of crossover support. It’s difficult to say whether or not this will disipate as people realize just how liberal he is, but my guess is, as usual, people vote for the man before the ideas. I think Obama is going to turn out to be quite electable.
January 21st, 2007 at 6:03 pm
I think that, given the recent election and current public opinion about the Bush administration, it should be no surprise that more people identify as Democrats than Republicans when the pollsters call. They’re most likely giving their honest opinion, at this point in time, about the head-to-head matchups.
January 21st, 2007 at 6:11 pm
GT,
Yes that is how it works. This is exactly how the MSM spins the numbers to get whatever result they wish to. They can choose to weight the poll however they wish (I would suggested that they use a party id that has the norm for the past 5 or so election which is either: even, Dems +2, or GOP +2). However, they chose to weigh it Dems +11.
But you are free to keep drinking the Kool-Aid all you like GT.
January 21st, 2007 at 7:31 pm
“I think people are missing the point here. There was an election 2 months ago. The electoral breakdown was 38D/36R.”
Right. And, based on that result, the semi-apathetic voters who only bother to vote in Presidential election years are taking cues from the result of that election — and they are taking cues not only in candidate preference, but party affiliation as well. Perhaps semi-apathetic Republican leaning independants are now identifying themselves as independants, while the Democratic leaning independantsare identifying fully with their party.
Also, I am a bit skeptical of the 2006 exit poll methodology, if that is in fact the party breakdown. Remember, the pre-election 2006 polls were generally on target, but they predicted a much more lopsided Democratic turnout. The only explaination I see is that either the pre-election polls incorrectly identified the party affiliation of the likely voters, or the exit pollsters did.
Don’t get me wrong, this poll is rigged, in a sense — but it isn’t rigged by Newsweek, so far as I can tell. Rather, the Democrats are enjoying a bounce in the polls — and one that will likely last through spring at the least — that is inevitable given their clear win in the polls in 2006. There is no reason to panic, but there is no reason to cry foul over the polling methodology either.
January 21st, 2007 at 10:13 pm
Wow. A widely unknown black man, with little if any experience and a middle and last name that matches or approximates two of our great enemies, is within 6 points of the god-like hero of 9/11 – within a few days of announcing an exploratory committee. Stunning!