PRINCETON, NJ -- A new USA Today/Gallup poll of likely New Hampshire Democratic primary voters shows Sen. Barack Obama and Sen. Hillary Clinton in a dead heat, each with 32% support. Former senator John Edwards is a solid third at 18%. The remaining candidates -- including Gov. Bill Richardson (8%), Sen. Joe Biden (4%), Rep. Dennis Kucinich (3%), and Sen. Chris Dodd (1%) -- are in single digits.
The poll was conducted Dec. 17-19, interviewing a sample of 750 New Hampshire registered voters who plan to vote in the Jan. 8 Democratic primary, including 510 who can be considered "likely voters." The New Hampshire primary is the critical second test in the nomination process. Polls of voters in Iowa, which holds the first contest, currently suggest a tight three-way race between Clinton, Obama, and Edwards.
Fifty-six percent of all likely New Hampshire primary voters (and 61% of those who have a definite first choice at this point in time) say they are certain to support their chosen candidate, meaning just under half can be swayed in the remaining days before the Jan. 8 contest. At this point, Clinton and Obama can each claim the definite support of about 1 in 5 likely voters.
Clinton, as she does nationally, fares better among women than men. Thirty-six percent of female likely voters currently support Clinton compared with 25% of men. Obama, in turn, does slightly better among men (35%) than women (30%).
Clinton's support is greater among registered Democrats than among those who have not declared a party affiliation. Among registered Democrats, the ballot is 36% Clinton, 29% Obama. Among undeclared voters, 36% support Obama and 28% Clinton.
Perceptions of the Candidates
Candidate images are beginning to crystallize in voters' minds, and these images to a large degree reflect the candidates' campaign messages. Clinton has emphasized her many years of experience in Washington during the campaign, while Obama has made the case that he is most likely to bring a fresh approach to the federal government.
New Hampshire Democrats seem to be buying Obama's campaign argument that he can best bring "change" to Washington. Specifically, 42% choose Obama as the candidate who has "new ideas that would help solve the country's problems." That is more than double the proportion who attribute this characteristic to Clinton (19%) and Edwards (20%).
Voters also view Obama as the candidate most "in touch with the average voter," by a 37% to 28% margin over Edwards. Clinton places third on this measure at 19%.
Obama has only small advantages over Clinton and Edwards in terms of sharing voters' values and standing up for what he believes in.
Clinton's campaign theme of "competence" is also resonating among New Hampshire voters, both in terms of her ability to negotiate the Washington policy process successfully and to win elections.
Half of New Hampshire Democratic voters say Clinton is the candidate most able to "get things done in Washington," with 18% saying this most applies to Obama and 14% to Edwards.
New Hampshire likely primary voters view Clinton as having the best chance of beating the Republican, with nearly half (47%) saying she does, while 26% say this about Obama, and 17% of Edwards. Perceived candidate viability was a key driver of Iowa and New Hampshire voting decisions in 2004 according to exit polls.
Three issues are clearly at the tops of Democratic voters' minds -- Iraq, healthcare, and the economy. When asked to choose among these issues plus illegal immigration, terrorism, and taxes as most important to their vote, 35% choose Iraq, 27% healthcare, and 26% the economy.
Clinton has long been associated with the healthcare issue and is the clear leader over Obama and Edwards among voters who choose healthcare as their top issue. Obama and Clinton are more closely matched among voters naming the economy or Iraq as their chief concern, with Obama holding a slight edge.
It's possible that issues could play a more crucial role in determining New Hampshire voters' decisions in the 2008 campaign than in 2004. When Gallup asked during the 2004 campaign whether issue agreement with the candidate or the candidate's chances of beating the Republican was more important to their vote, Democrats were evenly divided. In the current poll, voters say issue agreement is more important by a better than 2-to-1 margin. If voters' priorities shift during the final days of the campaign to a distribution similar to the 2004 campaign, Clinton stands to benefit given her status as the candidate viewed mostly likely to defeat the Republican in November.
Issues seem less important a factor when pitted against candidate leadership ability. Sixty percent say a candidate's leadership skills and vision will mater more than their issue positions, while 33% say issues matter more. New Hampshire Democratic voter preferences do not vary by whether candidates believe leadership or issues are more important.
The current national polls of Democrats' nomination preferences seem to suggest that Clinton should handily win the Democratic nomination. For example, in the latest Gallup national poll, 45% of Democrats (and Democrat leaners) nationwide rate Clinton as their top choice compared with 27% for Obama.
But as polls of New Hampshire and Iowa show, Obama is seriously challenging her in these critical early contests. Edwards is also a legitimate contender in Iowa.
An upset win by either candidate in Iowa or New Hampshire could easily transform the race, just as Kerry's upset win in Iowa propelled him from a middle-of-the-pack candidate to the clear front-runner and eventual nominee.
The Gallup New Hampshire poll suggests that what happens in the final critical days before the New Hampshire primary could tilt the outcome in Clinton's or Obama's favor
These results are based on telephone interviews with a randomly selected national sample of 510 likely Democratic primary voters in New Hampshire, aged 18 and older, conducted Dec. 17-19, 2007. Interviewing was conducted via landline telephone (for all respondents with a landline) and cell phone (for cell phone only respondents).
The likely voter model assumes 60% of self-reported Democratic primary voters will turn out, which represents approximately 25% of New Hampshire adults.
For results based on this sample, one can say with 95% confidence that the maximum error attributable to sampling and other random effects is ±5 percentage points.
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.