AUGUSTA, Maine — As Mainers prepare to vote for the second time in three years on whether to legalize same-sex marriage, national polls and surveys show a steady shift toward acceptance of the concept.

Despite that apparent trend, no measure to legalize gay marriage has ever won at the ballot box. Since 1998, voters in 32 states have cast ballots on the issue, and all have either rejected legalized same-sex marriage, banned it or endorsed the definition of marriage as a union between one man and one woman.

In addition to Maine, voters in Maryland, Minnesota and Washington state will decide ballot questions related to same-sex marriage on Tuesday. Question 1 on the Maine ballot asks: Do you want to allow the state of Maine to issue marriage licenses to same-sex couples?

In November 2009, Maine voters, by a 53-47 percent margin, overturned a same-sex marriage law passed by the Legislature earlier that year. Polls show majority support for a yes vote on Question 1.

But Maine polls before the 2009 vote and pre-election polling on same-sex marriage in other states also indicated more support than when ballots were tallied, and Patrick Murphy, who conducted a Pan Atlantic SMS Group poll this year, raised concerns about contradictory responses on gay marriage from some of the 400 likely Maine voters who participated in that survey.

Nevertheless, Lanae Erickson Hatalsky of Third Way, a Washington, D.C.-based think tank that advocates for same-sex marriage, points to The Big Shift, an October 2012 report that parses 98 national surveys and 128,000 responses to demonstrate widespread growth in support for same-sex marriage since 2004.

The Third Way analysis echoes a July 2012 Pew Research Center report that reflects gradually increasing national support for same-sex marriage since 2001. With minimal exceptions, data available at shows a similar trend.

Hatalsky, who co-authored the Third Way report with Gregory B. Lewis, said in a phone interview Monday that nine rounds of research conducted during the past two years show that the shift derives primarily from people who formerly opposed gay marriage changing their minds on the subject. Increased acceptance of legalized same-sex marriage cuts across all demographic segments, even among the nation’s most conservative groups, the report states.

“Although different demographic groups have warmed at different speeds, every single group has moved in a positive direction since 2004. Even among evangelical Protestants, one of the groups most likely to oppose marriage for gay and lesbian couples, support grew by 8 points from 2004 to 2011,” the report states.

Even with that growth, the Third Way research shows that only one in every five evangelical Protestants supports allowing gay couples to marry.

Hatalsky and Lewis relied on state polls, focus groups, psychological interviews, in-depth clinical interviews and online ad tests — “the full gamut of qualitative research,” Hatalsky said — to gauge support for same-sex marriage by age, religion, political party, political ideology, church attendance, race, region, education and gender.

“The thing that was underscored by the data was that some people say that the movement on marriage is happening in certain sectors, but really it’s happening in every one of them,” Hatalsky said.

Carroll Conley of Protect Marriage Maine, which leads the campaign against Question 1, said that “Whenever people talk about momentum, it has never occurred by voters.” He pointed to a May 2012 vote that amended North Carolina’s constitution to ban gay marriage. The ban passed overwhelmingly, and “If you took out all the voters over 45, it still would have passed by 8 percent,” according to Conley.

“The number people hate is 32,” he said, referring to 32 state votes against gay marriage.

No demographic sampling’s attitudes grew more negative toward same-sex marriage, and the greatest movement in favor of allowing gay couples to marry occurred among those who identify themselves as political moderates, the Third Way report indicates. That group showed a 21 percent increase in support for gay marriage since 2004.

The rise in apparent acceptance of same-sex marriage among moderates aligns with a 15 percent increase among independent voters included in the Third Way research.

“There’s been a huge rise in independent voters, and a lot of them are the Ron Paul libertarian voters,” Hatalsky said. “They are less religious, more secular and against government intrusion. This is a new piece in the marriage discussion in the last four years.”

A growing national divide between Democrats and Republicans manifests itself on the same-sex marriage issue, as 58 percent of Democrats endorse gay marriage , compared with 26 percent of Republicans, according to the Third Way report.

That split can be seen in the parties’ official positions in Maine. The Maine Democratic Party platform supports civil marriage for gays and lesbians “and opposes legislation to deny these rights to any family.” The Maine Republican Party platform states that, “Marriage is an institution between a man and a woman.”

Lewis and Hatalsky posit that “the probability of supporting marriage goes up 0.8 percentage point by each year of birth (except for a small plateau between 1953 and 1961) — meaning a person born in 1981 is 16 points more likely to favor marriage than someone born in 1961.”

While younger voters in all demographic samplings showed greater likelihood of supporting gay marriage, the overall trend toward acceptance derives more from people changing their minds than from younger voters supplanting older voters in the electorate, Hatalsky said.

What makes people change their minds?

Hatalsky believes a strategic refocusing on the institution of marriage rather than on equal rights explains much of the reconsideration.

“There was an answer to the legal arguments: civil unions,” she said. “When you talk about importance of marriage in society and want to participate in that tradition, then people were much more open. … The older- and middle-aged people needed a little more help in understanding that gay couples want the same things from marriage that other people do: love, commitment and spending lives together.”

That argument serves as the centerpiece for this year’s campaign to pass legalized marriage in Maine, according to David Farmer, spokesman for Citizens United for Marriage, which is leading the push for a yes vote on Question 1. He is also a BDN columnist.

“Same-sex couples want to get married for similar reasons my wife and I got married,” Farmer said. “We wanted to stand before our friends and make the promise that we would stay together. It helps all marriages when we recognize that the institution is strengthened when loving couples want to get into it.”

Farmer said that same-sex marriage advocates in Maine have conducted more than 250,000 conversations with the state’s voters. “We’ve been told by people that they changed their minds. Once they have the conversation about why marriage matters, the level of support goes up pretty fast,” he said.

However it’s measured, support for same-sex marriage has never translated into victory on Election Day. In a 2010 study of polling for same-sex marriage questions between 1998 and 2009, New York University professor Patrick Egan estimated that pre-election surveys typically inflate support for same-sex marriage by 7 percentage points.

Egan’s research rejects some pollsters’ suggestions that the gap can be attributed to voter confusion or to a “social desirability bias” in which poll respondents say they support same-sex marriage so as not to appear prejudiced, then vote against it.

Acknowledging Egan’s premise, Hatalsky said Third Way counted all undecided responses as opposition to same-sex marriage in its analysis.

In an Oct. 17 Reuters report, Brian Brown, president of the National Organization for Marriage, a leading opponent of gay marriage, argued that pre-election polls are unreliable because the questions are biased.

“If it’s framed as fairness or equality, you see movement,” Conley, of Protect Marriage Maine, said. “If you ask the question of whether marriage should be a union of one man and one woman straight up, the needle doesn’t move as far. If you do polling about the consequences, then we see the numbers down to where they were in 2009.”

Hatalsky agreed that framing the question affects responses. “There’s probably about 10 different ways to ask the question,” she said. “For example, offering civil union in the question lowers support for marriage. Our study took all of the different ways to ask the question and weighted them on how they would push people in one direction to get the most conservative estimate of what people say on these polls.”

With the new data in hand, what will Hatalsky be looking for to see if Maine voters break the 32-vote losing streak for legalized same-sex marriage?

“There will have to be a lot of movement in the middle age category,” she said. “Whether those people change their minds will make the difference.”

Robert Long is a political analyst for the BDN.