Poll: Still Close in Florida, Bush by Three Points

GALLUP NEWS SERVICE

PRINCETON, NJ -- A new CNN/USA Today/Gallup poll in Florida shows virtually no change in the presidential contest since last month, with likely voters leaning toward President George W. Bush by three percentage points. Similar polls of likely voters in Iowa, Nevada, and West Virginia also show Bush with single-digit leads of six to nine points that are just below or at the polls' respective margins of error.

Former Vice President Al Gore won Iowa in 2000 by a mere 0.3  percentage point, but the current poll shows Democratic candidate John Kerry trailing Bush by six points among likely voters, 50% to 44%.

The other three states reported here were all won by Bush in 2000, though by closer margins than what the polls now show.

• Bush's 2000 margin in Florida was a mere 537 votes, not even 0.1 percentage point, while now the poll shows a 3-point advantage for Bush among likely voters and a 2-point margin among registered voters.
  
  
• Nevada went for Bush in 2000 by 3.5 percentage points, but Bush has a 9-point lead in the poll among likely voters. However, among the larger group of registered voters, Bush's lead is just two points in the poll -- well within the poll's margin of sampling error.
  
  
• Bush leads by six points among likely voters in West Virginia, and by nine points among registered voters. In 2000, he won the state by six points.

Q.4/4A (Kerry vs. Bush vs. Nader)

No Significant Change in Presidential Contest in Florida

Among likely voters in the three polls conducted in Florida in July, August, and September, Kerry's support has been steady at 46%, while Bush's vote has fluctuated slightly, from 48% to 50%. 

The consistency of the vote is also evident among registered voters, with Kerry receiving 44% to 45%, and Bush 45% to 49%.

 

	Kerry/ Edwards	Bush/ Cheney	Nader/ Camejo	Neither/other/ no opinion
	%	%	%	%
Likely voters
2004 Sep 18-22	46	49	2	3
2004 Aug 20-22	46	48	2	4
2004 Jul 19-22	46	50	1	3
Registered voters
2004 Sep 18-22	45	47	2	6
2004 Aug 20-22	45	45	3	7
2004 Jul 19-22	44	49	2	5

Florida Senate Race Tilts Toward the Democrats

Florida's Democratic Sen. Bob Graham is retiring this year, giving the Republican Party an opportunity to increase its margin in the U.S. Senate. But the poll suggests that task may be a challenge. The Democratic candidate, Betty Castor, leads her Republican opponent, Mel Martinez, by six points among both likely voters (51% to 45%) and registered voters (49% to 43%). Martinez was Bush's Secretary of Housing and Urban Development until he stepped down in December, after reportedly being encouraged by the Bush administration to run for the vacant seat.

If the elections for senator were being held today, which candidate would you vote for in your state? As of today, do you lean more toward -- ?

Other questions of Florida respondents show the following:

• One in four Floridians say they have personally suffered significant property damage or financial loss this year as a direct result of a hurricane hitting the area where they live. In August, 10% of all Floridians said they had personally suffered loss from either Tropical Storm Bonnie or Hurricane Charley.
  
  
• Today, 62% of all Floridians are worried about suffering significant property damage or financial loss from a hurricane, up from 52% who expressed that view last month.
  
  
• Perhaps as a consequence of the hurricane damage, just 43% of likely voters in Florida rate the economy in their state as either excellent or good, down from 54% who gave it that rating in July.

Survey Methods

Results are based on telephone interviews with 1,003 Florida adults, aged 18 and older, conducted Sept. 18-22, 2004. For results based on this sample, one can say with 95% confidence that the margin of sampling error is ±4 percentage points.

For results based on the sample of 843 registered voters in Florida, the maximum margin of sampling error is ±4 percentage points.

Results for likely voters in Florida are based on the subsample of 674 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 Florida, turnout is assumed to be 50% of the voting age population.

Approximately 6% of all interviews were conducted in Spanish.

Results are based on telephone interviews with 721 registered voters in Nevada, aged 18 and older, conducted Sept. 18-21, 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 Nevada are based on the subsample of 535 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 Nevada, turnout is assumed to be 50% of the voting age population.

Approximately 5% of all interviews were conducted in Spanish.

Results are based on telephone interviews with 775 registered voters in West Virginia, aged 18 and older, conducted Sept. 17-20, 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 West Virginia are based on the subsample of 619 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 West Virginia, turnout is assumed to be 50% 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.

Results are based on telephone interviews with 766 registered voters in Iowa, aged 18 and older, conducted Sept. 16-19, 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 Iowa are based on the subsample of 631 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 Iowa, turnout is assumed to be 65% 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.