For all your poll-watching needs
Jun. 26th, 2004 09:11 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinThe diligent Professor Pollkatz has been back in action for a while, with his true identity revealed! (Persons of anything other than an enraged-Democrat persuasion may be put off a bit by his tone, but as far as I can tell his data-collecting, analysis and presentation are scrupulously honest.)
Which gives me yet another occasion to say this: People confused or dismayed by week-to-week fluctuations in presidential job approval polls would do well, as always, to look at the big picture. Opinion polls really do indicate things, but to extract real meaning you need to aggregate a lot of data; the week-to-week fluctuations in a single poll, which are the basis for most news stories about polls, are almost certain to mislead.
Also, R. Chung's analysis page is really nice, especially the graphs of systematic bias patterns.
Pollkatz used to have historical Gallup comparison charts up too, but it appears that the Gallup Organization nixed them, and Radio Free Monkey's similar site is down. Suffice it to say that in job approval, Bush is currently doing slightly better than his dad at this point in 1992 and Jimmy Carter in 1980, but not by much; and I think he's actually doing about as badly as they did in approval-disapproval spread, as there are fewer people on the fence. In approval, he's about even with Ford in '76. Since this sort of poll started being taken, everyone who won had job approval trending up at this point, not down. Now, I wouldn't take this as ironclad law; '76 was actually pretty close, and Bush Senior arguably could have turned it around in '92. Attempts to project electoral votes based on current polling data show a remarkably 2000-like knife-edge balance. But things really don't look good for Bush; a winner really ought to be doing a lot better, with John Kerry holding back a bit, and a long way from the convention that would traditionally formally introduce the challenger and kick off the national campaign.
The laments I frequently hear about America's unshakeable love for Bush are simply counterfactual. Sometimes people complain that he should be hated much more than he is; but even Nixon on the eve of his resignation only got down to about 25% job approval, and it took several months of pounding over Watergate for him to get there. Reagan's astounding reelection landslide in 1984 was about 60-40 in the popular vote. The 90-10 split never happens.
Which gives me yet another occasion to say this: People confused or dismayed by week-to-week fluctuations in presidential job approval polls would do well, as always, to look at the big picture. Opinion polls really do indicate things, but to extract real meaning you need to aggregate a lot of data; the week-to-week fluctuations in a single poll, which are the basis for most news stories about polls, are almost certain to mislead.
Also, R. Chung's analysis page is really nice, especially the graphs of systematic bias patterns.
Pollkatz used to have historical Gallup comparison charts up too, but it appears that the Gallup Organization nixed them, and Radio Free Monkey's similar site is down. Suffice it to say that in job approval, Bush is currently doing slightly better than his dad at this point in 1992 and Jimmy Carter in 1980, but not by much; and I think he's actually doing about as badly as they did in approval-disapproval spread, as there are fewer people on the fence. In approval, he's about even with Ford in '76. Since this sort of poll started being taken, everyone who won had job approval trending up at this point, not down. Now, I wouldn't take this as ironclad law; '76 was actually pretty close, and Bush Senior arguably could have turned it around in '92. Attempts to project electoral votes based on current polling data show a remarkably 2000-like knife-edge balance. But things really don't look good for Bush; a winner really ought to be doing a lot better, with John Kerry holding back a bit, and a long way from the convention that would traditionally formally introduce the challenger and kick off the national campaign.
The laments I frequently hear about America's unshakeable love for Bush are simply counterfactual. Sometimes people complain that he should be hated much more than he is; but even Nixon on the eve of his resignation only got down to about 25% job approval, and it took several months of pounding over Watergate for him to get there. Reagan's astounding reelection landslide in 1984 was about 60-40 in the popular vote. The 90-10 split never happens.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2004-06-26 07:59 pm (UTC)