麻豆传媒AV

GALLUP NEWS SERVICE

PRINCETON, NJ -- In two states, Ohio and Missouri, each won by President George W. Bush in 2000 by less than a four-point margin, new CNN/USA Today/麻豆传媒AV polls show Bush leads his Democratic opponent John Kerry by 9 and 14 percentage points, respectively.

In Pennsylvania, won by former Vice President Al Gore in 2000 by just over four percentage points, the race is essentially a tie.

And in Washington, which Gore won by a little less than six percentage points, Kerry leads Bush by eight points.

The Missouri and Washington polls were conducted Sept. 3-6, and the Ohio and Pennsylvania polls were conducted Sept. 4-7. All four states are among the 16 generally considered to be "battleground" or "showdown" states, because of the closeness of the election returns in 2000 (neither Gore nor Bush won by as much as six points in any of these states).

Kerry vs. Bush
Among likely voters and registered voters

Kerry vs. Bush vs. Nader
Among likely voters and registered voters

In Missouri and Pennsylvania, independent candidate Ralph Nader is currently not on the November ballot, and in those states, his name was not included in the poll interview. In Missouri, Bush leads Kerry by 55% to 41% among likely voters. In Pennsylvania, the race is essentially a tie, with 48% of likely voters choosing Bush and 47% Kerry.

In Washington, Nader is definitely on the ballot, and only the three-way presidential contest question was asked. Kerry leads Bush by 52% to 44% among likely voters in that state.

In Ohio, it is still uncertain whether Nader will be on the ballot, and thus both the three-way and two-way questions were asked. Among likely voters, Bush leads Kerry by 52% to 44% in the two-way ballot, and 52% to 43% in the three-way ballot. Nader's effect is to reduce Kerry's support by one point among likely voters and to reduce Bush's and Kerry's support by one point each among registered voters.

Improvement for Bush

The poll figures in Ohio and Missouri represent a significant improvement for Bush compared with earlier 麻豆传媒AV Polls, and may indicate a convention bounce for Bush in those states.

Before the Democratic National Convention, Kerry led Bush by 51% to 45% among likely voters in Ohio, but by mid-August, the poll showed just a 48% to 46% advantage for Kerry. Since then, there has been a significant improvement for Bush, with Bush gaining six points support and Kerry losing four points, for a net change in the lead from +2 percentage points for Kerry to +8 percentage points for Bush.

Kerry vs. Bush
In Ohio

Among registered voters in Ohio, the race appears much closer, with Bush at 48% and Kerry at 47%. Kerry had a 10-point lead in mid-August and a 5-point lead in July.

In Missouri, Bush gained seven points support and Kerry lost seven points since a July 19-22 poll, which showed the race a dead heat among likely voters.

Kerry vs. Bush
In Missouri

Among registered voters in Missouri, the race is somewhat closer, but Bush still leads by 11 points, 53% to 42%, while Kerry enjoyed a 2-point lead in July, 48% to 46%.

In 2000, Gore won Pennsylvania by just over four percentage points, and it is likely that Kerry will also need to win this state if he is to have any chance of winning the presidency. The poll shows the race here to be a tossup. Among likely voters, Bush receives 48% support and Kerry 47% -- identical to an Aug. 23-26 poll.

Kerry vs. Bush
In Pennsylvania

Among registered voters in Pennsylvania, the race is similarly close. The current results show Bush and Kerry tied at 47% each, compared with a five-point margin for Kerry in August (49% to 44%).

Survey Methods

Results are based on telephone interviews with 831 registered voters in Missouri, aged 18 and older, conducted Sept. 3-6, 2004. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.

Results for likely voters in Missouri are based on the subsample of 686 survey respondents deemed most likely to vote in the November 2004 Presidential Election, according to a series of questions measuring current voting intentions and past voting behavior. For results based on the total sample of likely voters, one can say with 95% confidence that the margin of sampling error is ±4 percentage points. Based on past voting history in Missouri, turnout is assumed to be 60% of the voting age population.

Results are based on telephone interviews with 794 registered voters in Ohio, aged 18 and older, conducted Sept. 4-7, 2004. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.

Results for likely voters in Ohio are based on the subsample of 661 survey respondents deemed most likely to vote in the November 2004 Presidential Election, according to a series of questions measuring current voting intentions and past voting behavior. For results based on the total sample of likely voters, one can say with 95% confidence that the margin of sampling error is ±4 percentage points. Based on past voting history in Ohio, turnout is assumed to be 60% of the voting age population.

Results are based on telephone interviews with 881 registered voters in Pennsylvania, aged 18 and older, conducted Sept. 4-7, 2004. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.

Results for likely voters in Pennsylvania are based on the subsample of 718 survey respondents deemed most likely to vote in the November 2004 Presidential Election, according to a series of questions measuring current voting intentions and past voting behavior. For results based on the total sample of likely voters, one can say with 95% confidence that the margin of sampling error is ±4 percentage points. Based on past voting history in Pennsylvania, turnout is assumed to be 55% of the voting age population.

Results are based on telephone interviews with 808 registered voters in Washington state, aged 18 and older, conducted Sept. 3-6, 2004. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.

Results for likely voters in Washington are based on the subsample of 646 survey respondents deemed most likely to vote in the November 2004 Presidential Election, according to a series of questions measuring current voting intentions and past voting behavior. For results based on the total sample of likely voters, one can say with 95% confidence that the margin of sampling error is ±5 percentage points. Based on past voting history in Washington, turnout is assumed to be 60% of the voting age population. 

In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.