Sign in to comment!

Menu
Home

Politics

Power Play

Why Polls Overstate Support for Gay Marriage

“The share of voters in pre‐election surveys saying they will vote to ban same‐sex marriage is typically seven percentage points lower than the actual vote on election day.”

-- New York University Political Science Professor Patrick Eagan writing in a 2010 study on the inaccuracy of pre-election polls on referenda banning gay marriage.

Does the power of states to regulate marriage extend to the ability to prohibit people of the same sex from exchanging vows?

That is the question before the Supreme Court today, and proponents of same-sex marriage believe they are on the cusp of a sweeping victory in which justices will strike down not just California’s ban on the practice but those in 28 other states.

The argument, embraced by many politicians in the past two months, including President Obama, is that marriage is a human right and therefore it is no more acceptable to prohibit marriage between two men or two women than it would be to prohibit interracial marriage or marriage by the developmentally disabled, both of those bans having been disallowed by the high court.

This is the first chance for the Supreme Court to weigh in on the main relevant question about same-sex relations and the law: How much of homosexuality is a biological trait, as proponents say, “born that way” and how much is a preference based on one’s life experiences? And by what degrees?

(This case shows why the movement inside the gay community for at least a decade has grown intolerant about ideas of sexual preference. To win in court and claim the same high ground as the civil rights movement, it cannot be about what one prefers but rather an immutable biological imperative. The old thinking in that community about sexual preference existing on a continuum and love existing between individuals rather than genders is more romantic, but less legally useful.)

The spate of red-state Democrats, including the politically vulnerable Sen. Mark Begich of Alaska, and moderate Republicans, like Ohio Sen. Rob Portman, who have followed Obama into this rapid evolution have no doubt been heartened by poll numbers that show surging support for same-sex marriage.

And they’re right. No social issue since perhaps segregation has seen such a sudden shift in popular opinion. The latest FOX News poll found 49 percent of voters in support of legalizing gay marriage, up 17 points in a decade.

But 53 percent thought that states ought to be able to ban the practice, including, contradictorily, 30 percent who believed that same-sex marriage is a constitutional right.

While Power Play does not expect constitutional coherency on public opinion, especially on a question so fraught with controversy and amid such rapid attitudinal evolution, there would seem to be something else afoot here than a misunderstanding of federalism.

Voters have had 31 chances to ban gay marriage on the state level, and in 29 of the cases have opted to do so, even in some pretty liberal places, including in California, the state ban before the court today. And in every race but one, pre-election polls have underestimated support for the ban.

Republicans looking for ways to believe that pre-election polls were wrong in 2008 and 2012 talked a great deal about the so-called “Bradley effect,” a reference to the 1982 California gubernatorial race in which L.A. Mayor Tom Bradley, the first black major-party candidate for California governor, lost to Republican George Deukmejian despite Bradley holding solid leads in pre-election polls.

The effect of what academics call “social desirability bias” in that election and other races involving minority candidates has been hotly debated for decades, but all agree that in Bradley’s race some portion of respondents felt uncomfortable telling pollsters that they were opposed to a minority candidate. Once in the privacy of a voting booth, they expressed their true feelings.

Some folks are lying to pollsters.

Republicans looking for a way to disbelieve pre-election polls in 2008 and 2012 often cited this phenomenon, suggesting that President Obama’s support was overstated. Post-election analysis revealed little if any evidence of a social desirability bias. (If the election of an African-American president does not prove the diminution of racism in America, then the honest expression of opposition to him to pollsters certainly should.)

But when it comes to gay marriage, social desirability seems to be a pretty big deal. The success of gay marriage bans doesn’t just reflect the intensity of the opponents and their willingness to head to the polls, but also the private preferences of those who might publicly state their support for same-sex couples.

Consider the North Carolina referendum last year. While pre-election polls showed the stringent ban was likely to pass, it did not foresee the 20-point blowout. It certainly did not predict that at least 35 percent of Democrats would vote in favor of the ban, even as Obama won a symbolic primary victory in the state.

There is more to this than simply the difference between the difference between the electorate and the general population. Some folks are lying to pollsters.

Gay activists should take this dishonesty as flattery. Twenty years ago, there wouldn’t have been much social desirability in expressing one’s support for same-sex unions. Just as 20 years before the Bradley election, there wouldn’t have been as much desirability for expressing support for a black candidate.

The disparity between public expressions and private votes is unlikely to be resolved in favor of restricting same-sex marriage. Embarrassment over an opinion is likely a precursor of changing it.

But that does not mean that the path to real majority support for people of the same gender marrying or gays and lesbians being treated with the same protections afforded minority groups and the disabled will be a smooth one.

In fact, if the court decides, as same-sex marriage advocates wish, and strike down the state bans, the issue will flare across the nation as couples head to courthouses in places like Virginia, Nevada and Ohio to take out their marriage licenses. Those with misgivings, including ones unexpressed to pollsters, may not like what they see.

Politicians who are jumping aboard the same-sex marriage train, particularly Democrats like Begich and Missouri Sen. Claire McCaskill who hail from states with lots of socially conservative voters in the liberal party, may be in for a surprise come voting time.

And with so many outside groups, Republican candidates may benefit from these unspoken attitudes without having to say much on the subject. When Alaskans for Traditional Marriage or whatever the issue group is called starts hitting Begich on his newfound support for the practice, his Republican challenger need only quietly restate his or her opposition to changing the law.

The arc of history may be bending toward same-sex couples, but it won’t be a smooth line.

And Now, A Word From Charles

"What's amazing here, I think, is how small Cyprus is and how relatively small the problem is. The bailout total, that you mentioned, is about a quarter of Apple's cash-on-hand. I mean, this is one country that Apple could purchase. It could own the island and call it, you know, iCyprus, or something. And have all this cash left over."

-- Charles Krauthammer on “Special Report With Bret Baier.”


Chris Stirewalt is digital politics editor for Fox News, and his POWER PLAY column appears Monday-Friday on FoxNews.com. Catch Chris Live online daily at 11:30amET  at  http:live.foxnews.com.

Chris Stirewalt joined Fox News Channel (FNC) in July of 2010 and serves as digital politics editor based in Washington, D.C.  Additionally, he authors the daily "Fox News First” political news note and hosts “Power Play,” a feature video series, on FoxNews.com. Stirewalt makes frequent appearances on the network, including "The Kelly File," "Special Report with Bret Baier," and "Fox News Sunday with Chris Wallace.”  He also provides expert political analysis for Fox News coverage of state, congressional and presidential elections.