NEW YORK (Reuters) - In Maryland, Maine and Washington, voters appear to be warming to the idea of legal marriage of same sex couples, raising the likelihood that come Election Day at least one of those states will join six others that have approved gay marriage.
That is, if the polls are to be believed.
Gay marriage activists suspect voters - especially those who believe marriage should be only between a man and a woman - might be unwilling to voice their true feelings in polls.
They fear there may not be as much support as polls suggest for gay marriage initiatives, particularly since many believe those in the "undecided" column will wind up in the "no" column.
The November 6 election could mark the first time that voters themselves decide same-sex marriage should be legal in their state. While six states, as well as Washington, D.C., now recognize such marriages, the change was made either by state legislatures or the courts.
Josh Levin, the campaign manager for Marylanders for Marriage Equality, says he expects the vote to be "very close," despite a Baltimore Sun poll in late September that found voters support the state's ballot initiative 49 percent to 39 percent.
Other activists were similarly cautious, even in the face of a string of polls that show rising support for same-sex marriage. In Maine, a Critical Insights poll in late September found 57 percent approval for legalizing same-sex marriage, while a recent SurveyUSA poll in Washington found voters favoring the change 56 percent to 38 percent.
In Minnesota, where voters will consider limiting marriage to opposite-sex couples, an October 8 poll by Public Policy Polling found 49 percent of voters did not support the restrictions to 46 percent who did. Gay marriage advocates have an advantage in the state because blank ballots will count as "no" votes.
Still, advocates of same-sex marriage are not ready to celebrate, having learned a hard lesson in 2008 when California voters rejected same-sex marriage despite a series of polls suggesting public opinion was moving in the opposite direction.
New York University professor Patrick Egan looked at polling from 1998 to 2009 and found opposition to same-sex marriage was underestimated by seven percentage points.
"The best guess about how polls translate into election results is to add just about all the people who say they are undecided to the share of people saying they're going to vote against gay marriage," Egan said in an interview.
UNDECIDED VOTERS MAY TIP ISSUE
"The shorthand is, unless the pro-gay marriage side is above 50 percent in the polls, then they have reason to fear that come election day they might find themselves on the losing side," said Egan, whose research was published in 2010.
There was no clear reason for this, Egan said. While some suspect a "social desirability bias," where respondents try to avoid appearing homophobic, there is no evidence it exists here.
Jim Williams, Issue Polling Specialist for Public Policy Polling, agreed. "My guess would be that certainly a lot of folks who say they're undecided are not going to vote for it," he said.
So as not to be overly confident the measure will pass, Matt McTighe, the campaign manager of Mainers United for Marriage, said he assumes those in the "undecided" column will vote for the other side.
Brian Brown, the president of the National Organization for Marriage, the leading group opposing same-sex marriage, said the proof is in the numbers: more than 30 states have voted to restrict marriage to unions between a man and a woman.
Brown argues that most polling on same-sex marriage is unreliable because the questions themselves are biased: voters are less likely to favor limiting rights, while many favor the idea of promoting traditional values.
There are also issues relating to the wording on a ballot, which may be different than the question posed by pollsters, experts said.
In Maryland, the referendum to be shown to voters - which runs nearly 100 words - states that gay and lesbian couples will be allowed to "obtain a civil marriage license," and sets out accommodations for clergy and religious groups.
Derek McCoy, who heads Maryland Marriage Alliance, which opposes the referendum, said the language was a "transparent attempt by the Secretary of State to bias voters to be in favor of the legislation," according to the Washington Blade newspaper.
In Minnesota, the state's high court rejected language proposed by the secretary of state for the referendum's title: "Limiting the Status of Marriage to Opposite Sex Couples."
Instead, voters will see: "Recognition of Marriage Solely Between One Man and One Woman."
"In our polling, if we ask about gay marriage, we get much less support than if we ask about marriage equality. And if we put people in a frame to think about individual freedoms, we get more support typically. So a lot depends on what's in people's head when they're thinking about this issue," said David Redlawsk of the Eagleton Center for Public Interest Polling in New Jersey.
(For a factbox on same-sex marriage see ID:nL1E8LAFII)