One Week Later, Gallup Says Bush Approval Rating Has Fallen From 57% to 49%

by Steve

A week after the most recent Gallup Poll for CNN and the USA Today claimed that Bush’s approval rating shot up to 57%, using a sample that had a 9 percentage point advantage for the GOP over Democrats (37% GOP, 28% Democrats), Gallup came out with its own poll last Friday. This poll, not done for CNN or USAT, and not bull-horned through the media and seemingly lost in the late Friday news dump, shows that Bush’s approval rating plummeted in one week to 49%, with his disapproval rating now up to 48%. As yet, I do not have the party ID breakdowns from Gallup on the Friday poll, but I suspect they will show something less than a 37% GOP-28% Democratic breakdown.

How often is there a 16% swing in a public opinion poll in one week?

As I said, the media is silent on this poll. A look at the Yahoo and Google front pages this morning yielded no mentions of a new Gallup poll that showed such bad numbers for Bush. And of course neither CNN nor USAT blasted this poll across their websites this morning. I wonder why the corporate conservative media doesn't find this a newsworthy story? Go figure.

Two things are noticeable in the story from Gallup’s website: first, the analysis by Gallup about why Bush’s numbers are falling; and second, Gallup’s attempt to support these lower numbers for Bush by comparing the new results as being in line with other recent polls like the AP/Ipsos poll last week.

Gallup makes the point that whatever bump Bush got from the SOTU and the Iraq election quickly dissipated when the focus turned to domestic issues like Bush’s Social Security benefit reduction/privatization plan and his budget cuts. Again, we can argue whether Bush ever had a bump to 57% from the SOTU and the Iraq election in the first place, when you had a sample that contained nearly a hundred more Republicans than Democrats. But Gallup feels that this recent drop is legitimate and tied to the problems Bush is having with the public on Social Security and his budget cuts.

Furthermore, Gallup points out that Bush’s approval ratings on the economy have fallen to 45%, a figure not seen since June of last year, even as the poll indicated that the public is gradually feeling better about the economy. This suggests that Bush isn’t getting any credit for the economy, and is only catching the negative fallout from his Social Security benefit cuts and budget choices. That message will not escape the Republicans in Congress.

Secondly, I think it is interesting that Gallup would compare their results in this poll with other pollsters who also show that Bush’s approval ranges anywhere from the AP/Ipsos 45% level up to Fox’s 51% level. I don’t recall seeing Gallup compare itself to other pollsters for validation of their results, and wonder why Gallup sees a need to do so here. Even Rasmussen has Bush at a 50%-49% split.

If Gallup, of all firms, says that Bush’s approval/disapproval is 49%-48%, then there are major problems at 1600 Pennsylvania Avenue, no matter what Muckdog or others think of Bush as a political card player. And what does it say about Bush's second term when he already has a disapproval rating as high as 48% or 49%, after rolling out his main agenda for the next two years? How bad will it get from here on?

I’m sure that the Republicans on the Hill are asking this also.

Update: Thanks to Gallup's courtesy and quick response, I received the party ID breakdowns for their last poll, which was 35% Democrats and 34% Republicans. Yet last week it was 37% Republicans and 28% Democrats, and in mid-January it was 33% Republican and 30% Democrat. Gallup, like other pollsters adjusts its samples to reflect census and demographic factors in the population as a whole and doesn't weight for party identification because they feel that party ID is too volatile. These numbers would lend credence to that concern. But do you think that party ID changes this much in four weeks?

January 14-15
Republican 33%
Independent 36%
Democrat 30%

February 4-6
Republican 37%
Independent 35%
Democrat 28%

February 7-10
Republican 34%
Independent 30%
Democrat 35%

When all is said and done, what this recent poll tells us is that when the sample has an almost evenly divided number of Republicans and Democrats, Bush's approval/disapproval numbers are also almost evenly divided. There's no surprise there.