Gallup Final Pre-Election Poll: Near-Deadlock

Poll: Bush, Kerry in near-deadlock (USA Today)

President Bush and John Kerry are in a near-tie in the final USA TODAY/CNN/Gallup poll before the 2004 election — a strong sign that this race, which has stayed close since Kerry became the likely Democratic nominee in March, will remain that way when voters go to the polls Tuesday.
Bush gets 49% and Kerry gets 47% among likely voters in the poll, which was conducted Friday-Sunday. Three percent offered no opinion. The poll has a margin of error of 2 percentage points, meaning Bush does not have a clear lead. Ralph Nader failed to break the 1% threshold, as all other candidates as a group drew a single percentage point of support. The poll used a sample of 2,014 national adults — a larger sample than past polls, which reduces the margin of error. In addition, in this final poll, Gallup used a statistical model to allocate undecided voters to the candidates. Using that model, the race is in a 49%-49% tie, with Nader getting 1% and all other candidates also receiving 1%.

The survey also is closer than the last USA TODAY/CNN/Gallup poll of a week ago, which gave Bush a 51-46 advantage.

The closeness of the 2004 election also is reflected in a number of USA TODAY/CNN/Gallup polls in so-called battleground states. Polls of likely voters in those states were as follows:

̢ۢ Florida: Kerry with 49%, Bush with 46%.

̢ۢ Ohio: Kerry 50%, Bush 46%.

̢ۢ Pennsylvania: Bush 50%, Kerry 46%.

̢ۢ Iowa: Bush 48%, Kerry 46%.

̢ۢ Minnesota: Kerry leading Bush, 52%-44%.

̢ۢ Wisconsin: Bush leading Kerry, 52%-44%.

All state polls concluded Sunday and had an error margin of plus or minus three percentage points.
The swing in the Florida results are perhaps the most dramatic. There have been six USA TODAY/CNN/Gallup polls in the state since mid-July, and this is the first one to give Kerry an advantage. However, that gap is still within the margin of error. The poll also reflected some changes in the way presidential elections are held. Seventeen percent of registered voters in the survey said they already had cast a presidential ballot in the election, and another 4% planned to do so before Election Day.

In other results from the poll:

• Bush had a 51-46 approval/disapproval rating from likely voters, with 3% expressing no opinion. That keeps him above the crucial 50% that often spells trouble for incumbents, but it is a three-point drop from a survey a week ago. But when likely voters were asked if they were satisfied or dissatisfied with the nation’s direction, 46% said they were satisfied and 52% were dissatisfied. Fifty-three percent of likely voters said they had a favorable opinion of Bush and 45% said it was unfavorable; Kerry’s numbers on the issue were 51-46.

I’m skeptical of Gallup’s state polls, which seem outliers, but Gallup is the granddaddy of them all and thus the one that’s used for historical purposes.

FILED UNDER: 2004 Election, Public Opinion Polls, , , , , , , , , , , , ,
James Joyner
About James Joyner
James Joyner is Professor and Department Head of Security Studies at Marine Corps University's Command and Staff College. He's a former Army officer and Desert Storm veteran. Views expressed here are his own. Follow James on Twitter @DrJJoyner.


  1. McGehee says:

    Bush had a 51-46 approval/disapproval rating from likely voters … But when likely voters were asked if they were satisfied or dissatisfied with the nation’s direction, 46% said they were satisfied and 52% were dissatisfied.

    The latter question might be ambiguous in a lot of respondents’ minds. They may be dissatisfied over things for which they don’t blame Bush — such as the vitriol of political messages during this campaign, or the blatant partisanship of CBS’ abortive “Bush AWOL” memo story and the like. Or even the threat of the courts imposing a redefinition of marriage over the expressed objections of the electorate.

    It might have been more useful for the question to be posed in more specific, Bush-centric terms.

  2. dw says:

    The “nation’s direction” question is a pretty solid pollable because it’s an unbiased aggregate question.

    But I could ask them to rephrase it for you if you wish.

    “Do you believe the country is going in the right direction or the wrong direction?”

    A. Right direction

    B. I am a homosexual French-speaking Communist appeasing flip-flopping abortion doctor who thinks Hanoi Jane is a MILF and wants to bring about the downfall of America by banning Krispy Kreme and encouraging teenage girls in southeastern Oklahoma to “go to the bathroom together” if you know what I mean and I think you do

    C. Don’t know, because I’m a wishy-washy brainless jellyfish who drives with my turn signal on and will be the first against the wall when the UN comes in their black helicopters

    Would this work better for you? Let me know. I have a conference call with Zogby tomorrow afternoon.

  3. McGehee says:

    The “nation’s direction” question is a pretty solid pollable because it’s an unbiased aggregate question.


    But it’s not something you can really use to extrapolate dissatisfaction with anything in particular for that very reason. The tendency is to try to reconcile majority approval of Bush with majority dissatisfaction with the nation’s direction, as though they are both concerned with the shiny red fruit — when I think what we’re both saying is, that ain’t so.

  4. dw says:

    On review, I don’t think I should have used “unbiased” there, because the question is biased (albeit a very low level on the surveyor’s side). I meant “simple,” or rather, “simplistic.”

    Think of it like baseball — “right direction” is like batting average. BA is a simple, binary statistic (you either get a hit or you don’t) that people use as a measuring stick because it’s consistent over time. Ichiro’s .372 is fundamentally comparable to George Brett’s .390 in 1980. Any good sabermetician will tell you, though, that BA is an awful stat because it doesn’t include XBH or BB. It is, though, a consistent measuring stick.

    So, I don’t think it’s a bad question to ask. You’re right — “right direction” is in the eye of the beholder, and one could have issues with the social direction of this country and still approve of the president’s policies. It’s a useful yardstick, though. And a six point differential is actually pretty low — I seem to recall Clinton having a 10+ point differential in ’96. The only reason it’s interesting this time is because Bush’s approval rating is hovering at 50% and “conveys divisiveness” where Clinton’s was hovering at 60%.