ABC Poll: Second Debate Changed Few Minds
ABC News Poll Shows Debate Had Little Impact on Voter Perceptions
Analysis By
BRIAN HARTMAN
Oct. 11, 2004— The second presidential debate left voter perceptions largely unchanged: John Kerry didn't further advance the gains he scored in the first debate, nor did George W. Bush either cede or reclaim ground.
Kerry remains more competitive in terms of personal popularity than he was before debate No. 1, an important improvement for him. But Bush continues to hold the upper hand on leadership, clarity and the issues of Iraq and terrorism.
|
| Sampling, data collection and tabulation for this poll were done by TNS. |
The race between them remains close: Fifty percent of likely voters prefer Bush, 46 percent Kerry, one percent Ralph Nader. These numbers have been quite stable, with Bush between 49 and 51 percent, and Kerry between 45 and 47 percent, since just before their first debate.
Movables
There's some room to move: Fifteen percent of likely voters say their minds are not definitely made up. But just 6 percent are either truly undecided or say there's a "good chance" they may change their minds.
There are essentially equal numbers of movables on each candidate's side — meaning they share the same opportunity, and the same vulnerability, among impressionable voters. That's what makes the debates potentially influential; the candidates meet Wednesday in their third and final debate, focused entirely on domestic issues.
Those issues could be fertile ground for Kerry, and they're of special interest to movable voters. movables are more likely to cite either the economy or health care as the most important issue in their vote — issues on which Kerry is more competitive — and less likely to cite terrorism or Iraq, issues on which Bush is stronger.
 Most Important Issue |
||
|
|
||
| Issue | Movable Voters | Definite Voters |
|
|
||
| Economy and Jobs | 28% | 23 |
|
|
||
| Health Care | 24 | 9 |
|
|
||
| War in Iraq | 18 | 25 |
|
|
||
| Terrorism | 15 | 25 |
|
|
||
| Education | 5 | 3 |
|
|
||
|
||
But as important as appealing to movables on the issues is the challenge of getting them to turn out on Election Day. They tend to be following the race less closely.
DEBATE
One reason the second debate didn't shake up perceptions is the view that the candidates battled to a draw. Among all likely voters, 35 percent say Kerry won, 32 percent Bush. It also was a draw among debate viewers Friday night.
After the first debate, by contrast, viewers gave Kerry the win by a 9-point margin, and in interviews over the next two days, likely voters more broadly said he'd won by a much wider 52-23 percent.
Who Won the Debate Among Likely Voters |
||
|
|
||
| Candidate | Second Presidential Debate | First Presidential Debate |
|
|
||
| Bush | 32% | 23 |
|
|
||
| Kerry | 35 | 52 |
|
|
||
| Tie | 16 | 15 |
|
|
||
| No Opinion | 17 | 10 |
|
|
||
|
||
Gender
Bush's support relies on men, and particularly on married men. Men overall prefer Bush by a 16-point margin, 56-40 percent. And men are 10 points more likely than women to call Bush's top issue, terrorism, the most important issue in their vote.
Married men, in particular, are more likely to be Republicans and more likely to support Bush. And terrorism is a much higher concern among married men than it is among either women or single men. (Among people who call terrorism their top issue — men and women alike — nine in 10 support Bush.)
Most Important Issue |
||
|
|
||
| Issue | Men | Women |
|
|
||
| Terrorism | 28% | 18 |
|
|
||
| War in Iraq | 21 | 27 |
|
|
||
| Economy and Jobs | 24 | 23 |
|
|
||
| Health Care | 9 | 13 |
|
|
||
| Education | 3 | 4 |
|
|
||
|
||
In vote preferences, being single — and the demographic differences associated with single people — is as strong a factor as sex. Single men and women alike prefer Kerry over Bush — single men by 58-39 percent, single women by 57-38 percent.
Married women split about evenly. And married men, by contrast, prefer Bush by 63-33 percent.
Methodology
This poll was conducted by telephone Oct. 7-10, 2004, among a random national sample of 2,409 adults, including 2,026 registered voters and 1,601 likely voters. The results have a 2.5-point error margin for the likely voter sample. Sampling, data collection and tabulation are conducted by TNS of Horsham, Pa.
See previous analyses in our Poll Vault.
Much better sampling and timeframe than the recent Zogby/Gallup silliness...