McCain and Mike Huckabee are neck and neck heading into Saturday's Republican primary in South Carolina, where the outcome could hinge on a bloc of undecided evangelical voters, according to a new McClatchy-MSNBC poll.
The survey, by Mason-Dixon Polling & Research, found a battle between McCain, an Arizona senator, and Huckabee, a former Arkansas governor, for the lead. It also revealed a close struggle for third between former Massachusetts Gov. Mitt Romney and former Tennessee Sen. Fred Thompson.
As in earlier contests, the margin is close, the voters are unsure and the campaign is volatile. Almost 1 in 10 likely voters said they were still undecided, and one-third of those who did express support for candidates said they might change their minds in the final hours.
The biggest bloc of undecided voters are evangelical Christians.
-- McCain, 27 percent.
-- Huckabee, 25 percent.
-- Romney, 15 percent.
-- Thompson, 13 percent.
-- Rep. Ron Paul of Texas, 6 percent.
-- Former New York Mayor Rudy Giuliani, 5 percent.
-- Rep. Duncan Hunter of California, 1 percent.
-- Undecided, 8 percent.
The poll also found that the economy is tops on the minds of voters in South Carolina, as it was in Michigan. Voters who said the economy was the most important issue split between McCain, 29 percent, and Huckabee, 24 percent, with 17 percent going for Romney.
McCain draws his strongest support from those older than 50, men, non-evangelicals, non-Republicans and those looking for a leader who can keep the country safe. He's campaigned as a war hero who knows how to lead the country during wartime, appealing to veterans in a pro-military state.
He had a 2-1 advantage over Huckabee among non-Republicans, a nearly 4-1 advantage among non-evangelicals and a 2-1 edge over Huckabee and Romney among voters who say the quality they seek most in a candidate is leadership.
Huckabee draws his greatest support from voters looking for a candidate who shares their values, where he had a 2-1 edge over Thompson, a better edge over Romney and a nearly 3-1 advantage over McCain.
Huckabee has campaigned as a devout Christian who'd oppose abortion and gay marriage but also as an economic populist who'd look out more for working people than corporate bosses.
He had an edge among evangelical voters, but not the lopsided margin he's had in other states, winning 33 percent of them to McCain's 20 percent, Thompson's 15 percent and Romney's 13 percent.
"It's a little more spread around here," said Brad Coker, the managing partner of Mason-Dixon.
The poll suggests that evangelicals remain key swing voters heading into the voting. There were more undecided evangelicals, 11 percent, than non-evangelical voters, 4 percent, for a ratio of 3-1.
More people call themselves evangelical Christians in South Carolina than in most other states, including Iowa, where Huckabee won most of them. That's partly because there are more of them in South Carolina.
But it's also a matter of culture, Coker said, in a Southern state where more people are likely to say they're born-again Christians. "You get more moderate conservative mainstream Protestants calling themselves evangelicals."
And they might be voting on issues such as the economy or national security rather than traditional "values issues," such as abortion. A deluge of ads blasting Huckabee for raising taxes in Arkansas could be splintering his support among evangelicals.
Also, Huckabee has more competition in South Carolina for social conservatives than he's had in other states, mainly from Thompson.
How those remaining evangelicals vote could well determine the outcome.
"It's a tossup," Coker said. "If Huckabee pulls this out, it will be because the rest of the evangelicals went for him. If McCain pulls this out, it will be because they went on other issues like national security."
HOW WE POLL
The McClatchy-MSNBC Poll is a snapshot of voter opinion at the time it was conducted. It isn't a prediction of how people will vote on Election Day.
The Mason-Dixon poll of 400 likely Republican primary voters in South Carolina was conducted by telephone Monday through Wednesday. Those interviewed were selected by a random variation of telephone numbers from a cross-section of telephone exchanges. That means that anyone in the state with a phone line had the same odds of being called as anyone else, except for people who use only cell phones. Cell phone numbers aren't in the exchanges.
The margin of error was plus or minus 5 percentage points. That means that 95 percent of the time, the correct numbers could be as many as 5 percentage points above the poll's findings or as many as 5 percentage points below them. The remaining 5 percent of the time, the correct numbers could vary even more.
The sampling margin of error doesn't include other variables that could affect results, including the way questions are worded or the order in which they're asked.