USA Today/Gallup Poll and New York Times/CBS News Poll

(Note: This "Gallup Guru" blog is posted on USA Today's Web site.)

Two different headlines Monday in The New York Times and USA Today about the impact of the Jeremiah Wright controversy came to different conclusions about Jeremiah Wright and Barack Obama. The Times' headline: "In Poll, Obama Survives Furor; but Fall Is the Test," while USA Today headline: "Flap over pastor pulls Obama down, poll finds."

The USA Today story is based on our Thursday through Saturday USA Today/Gallup poll, and certainly in my opinion the write-up and headline are accurate representations of our data. I'm less familiar with the latest New York Times/CBS News poll, but I have gone over the questions they asked and looked at the basis for their conclusions, and I can say that their write-up appears to be an accurate representation of what their poll showed.

Basically, this whole arena of trying to ascertain the impact of a particular event or occurrence on a presidential campaign is one of the trickier areas in which pollsters get involved, in my experience. Part of the issue becomes one of the particular questions asked, part of it the answers obtained, and part of it a matter of emphasis.

In this particular situation, there are two types of survey indicators that help provide an indication of the impact of the Wright flap: First, basic indicators of candidate images and ballot positioning. Second, direct questions asking about the impact of the Wright controversy.

In terms of the first type of indicators, one of the reasons why the two papers may have headlined different conclusions is the simple fact that the two polls came up with different estimates of the positions of the candidates in the minds of the underlying population of Democrats. The USA Today/Gallup poll found Hillary Clinton ahead of Obama among Democrats, while The New York Times/CBS News poll found Obama strongly ahead of Clinton. That single finding helped set the tone for the news coverage. Additionally, the USA Today/Gallup poll found that Obama's unfavorable rating had increased at least marginally, while the New York Times/CBS News poll did not.

The explanations for these differences in estimates of candidate positions in this hot political environment are complex, but it's certainly not unprecedented. I believe, and our Gallup tracking data confirm, that the positions of the two candidates can and are undergoing sometimes daily shifts as the campaign grinds on in these days before Tuesday's voting. Gallup tracking had Clinton ahead of Obama last week by four points at one point (not far off from what the USA Today/Gallup poll showed a few days later), but by Monday the tracking had come back to an Obama lead. Because there is no Republican race to capture news coverage space at the moment, the Democratic race dominates news coverage, with enough fascinating swings and issues to keep things generally up in the air. Voters are highly tuned in, and I think this all makes it more likely than usual that polls conducted at roughly the same time derive different estimates of the candidates' positionings.

It's worth noting that when the New York Times/CBS News poll and the USA Today/Gallup poll measured something a little more stable at the moment -- President Bush's job approval rating -- they both came up with exactly the same estimate, 28%.

Indeed, both polls also found the same direction of shift among Democrats in response to the question about which candidate would do the best job against John McCain.

The USA Today/Gallup poll showed that Democrats had shifted around to the point where they think that Clinton has the better chance of beating McCain in the fall. The New York Times poll found that Democratic primary voters are roughly split on this question of which candidate would do best in the fall, although this actually represented an increase in views that Clinton has the best chance against McCain compared to the previous week's Times/CBS poll. In fact, the Times write up went back further in time and noted: "In February, 59% called Mr. Obama the stronger candidate [against McCain], compared with 28% who named Mrs. Clinton. In the latest survey, the two were essentially tied."

There were other similarities in the findings of the two polls.

The New York Times asked if the Wright association had made voters more likely or less likely to support Obama, with 9% saying more, 15% less and the rest no difference. Among all Democratic voters in the USA Today/Gallup poll, it's about the same, with 19% of Democratic voters saying that the Wright flap made them less likely to want to vote for Obama.

The USA Today analysis emphasized that "the survey makes it clear the Wright controversy has grabbed public attention." The Times/CBS survey also found that 75% had heard or read a lot or some about the controversy, very similar to the percentage of USA Today/Gallup respondents who said that they had followed the controversy very, somewhat or not too closely -- excluding those who said they had followed the controversy "not at all."

Both polls found that those with an opinion about Jeremiah Wright were strongly unfavorable.

On a key question about the association between Wright's comments and Obama's beliefs, the USA Today/Gallup poll found that 21% of Americans believe that Obama agrees with Wright. The Times found that 43% of Americans say that Wright's comments have had a lot or some influenced on Obama's political beliefs, and 58% say Wright has had influence on Obama's spiritual beliefs.

Basically, both polls found that a minority, but in some instances a substantial minority, of Americans believe that the Wright controversy influenced their thinking. Both the USA Today and The New York Times write-up of their polls' results pointed out correctly the instances in which the data seemed to support more major change and those in which it did not.

All in all, I think that the careful reader comes away with a better understanding of what is happening by reading either or both stories. The fact of the matter is that the two polls had different estimates of where the candidates stand at the moment, which is perhaps a perplexing issue to many readers, but one of the facts of life when polling in a churning election environment. The headlines more than anything else followed this finding, and as such reflected accurately what their polls showed.