New York:
Meet the undecided voter.
Sometimes seen as the great prize of U.S. electoral politics, this elusive citizen is more likely to be female, white, lacking a college education and earning less than $25,000, according Reuters/Ipsos polling data accumulated over the course of 2012.
The "undecideds" make up roughly 6 percent of the electorate, with a slightly higher concentration in an Upper Midwest region including Ohio and Wisconsin, swing states that could determine the outcome of a close election.
Even though such voters are expensive to win over and perhaps less likely to make it to the polls, they continue to capture the attention of campaigns and political analysts for their potential to tilt the result in a tight race.
After a long campaign season, the undecideds have also earned scorn from some pundits and humorists, incredulous that anyone could remain uncertain after months of campaign advertising and media coverage. "Low-information voter" is one of the more polite names attached to this group.
"This year at the debates we should skip that thing where the undecideds dial in their reaction to every little moment," remarked television host Bill Maher, "and instead hook up the dial to their foreheads to see if there is any measurable brain activity."
Not all that sureCalling the satire unfair, Mr Romney pollster Neil Newhouse refers to the demographic tendencies found in the Reuters/Ipsos poll as "Walmart moms." Fifty-four percent of undecided voters are women.
"Overwhelmingly they are concerned with their family budgets, not the national budget," Mr Newhouse said. "They are concerned about putting food on the table or gas in the tank. They haven't tuned into the campaign yet because they are struggling with their daily lives. They are living one paycheck away from going off the financial cliff."
Among poll respondents who answered 10 on a 1-to-10 scale of their intention to vote, the undecided group was more likely to chose "none," "other," "don't know" or "refuse to answer" on a broad range of questions about the presidential candidates and national political issues.
On questions about the budget deficit, same-sex marriage, healthcare and immigration, this same subset was twice as likely as all likely voters to answer "unsure" when asked about their personal views.
More of them believe the country is on the wrong track - 75 percent, compared to 60 percent - and they are much more likely to respond "neither" when asked whether Mr Obama or Mr Romney is the more eloquent, Presidential, or likeable candidate.
"I just don't feel like either of them are trustworthy," said Helen Krueger, 54, an undecided voter and mother of seven from Grants Pass, Oregon, who voted for John McCain in 2008. "I normally make up my mind and know who I'm going to vote for. This time I need to do more research."
Mr Obama, she said, has done a "lousy job." Mr Romney? "I don't think we should be taxing the medium class and low class and letting the rich get by without having to pay (taxes), and that's what I feel he wants."
Ms Krueger said she receives three or four automated phone calls a week from the Romney campaign. She listened the first time. Now she just hangs up.
Lynette Povsha, 39, another undecided voter from Missoula, Montana, has been out of work three years and is ignoring her medical bills. She said neither candidate offers her any hope.
"The economy keeps getting worse. My electricity's been off for months. I don't qualify for (public assistance). I'm a white woman at 40 with no kids," said Povsha, who voted for McCain in 2008 and has yet to be contacted by the campaigns this year. "I don't see what either of these presidential wannabes can do for me. I might put Mickey Mouse down for President."
DeEntre Thompson, 38, a school bus driver from Columbus, Ohio, said he doubted either Mr Obama or Mr Romney could win his vote. He said he voted for Bill Clinton twice, George W. Bush twice, and "reluctantly" for McCain in 2008.
"With the two clowns that are running, I say 'undecided' because I don't have a third option," Thompson said, lamenting the lack of a strong third-party candidate. "If one of them showed up at my door and paid all my bills, he could sway me. Anything short of that, no."
Worth the effort?
Experts debate whether it is more cost-effective to court undecided voters or spend the time and money getting loyal supporters to the polls.
"What's most important in this race is which party is most able to motivate its base," said Andrew Smith, director of the University of New Hampshire Survey Center. Still, "it could be worthwhile for the campaigns to go after (undecided voters) because in several states it's a very close race, and it could be decided by one or two percentage points."
When Ipsos looked at current undecided respondents who voted in the 2008 presidential race, the undecideds were slightly more concentrated in a region the Census Bureau defines as the East North Central, comprised of Illinois, Indiana, Michigan, Ohio and Wisconsin. Nationally, 76 percent were white compared to 72 percent for all likely voters.
The Romney campaign is making a special effort to woo the undecideds in swing states with phone calls, direct mail, email and one-on-one contact, Newhouse said.
"They sound like voters who are up for grabs. These are encouraging demographics for us," he said, even though Romney's overall standing among women in the Reuters/Ipsos poll is 33 percent.
The Obama campaign has an extensive ground organization that emphasizes neighbor-to-neighbor contact with undecided voters.
That kind of personalised contact is the most effective way to mobilize the undecideds, said Diana Owen, a political science professor at Georgetown University, but it is also expensive.
"That being said,"Mr Owen said, "it's never good to write off a sizeable constituency. As we saw in 2000, it takes very few votes to switch the outcome of an election."
The poll has surveyed nearly 100,000 voters since January 1. This article cited data from August and September, representing a pool of more than 17,000 likely voters, and distinctions made about undecided likely voters versus all likely voters were based on a difference of at least three percentage points.
The Reuters/Ispsos database now available to the public can be searched to get a detailed look at the undecideds and other demographic groups.
© Thomson Reuters 2012