A poll question I hate

As I noted here years ago, there are few issues on which public opinion has changed more rapidly than gay rights. Forty years ago, in the University of Chicago’s General Social Survey (GSS), 72 percent condemned “sexual relations between two adults of the same sex” as “always wrong.” For the next 15 years that figure bounced between 67 percent and 75 percent. Then, rather suddenly, after 1988, it began a precipitous fall, declining almost 30 points over the ensuing 22 years, dropping just under a point and a half a year on average.


Marriage remained the highest hurdle. As recently as 1988, a mere 11 percent agreed that “homosexual couples should have the right to marry one another.” By 2010 (the latest GSS data available), support increased to 46 percent — an average growth rate of over a point and a half a year. Just six years after the first state began granting marriage licenses to gay couples, a plurality of Americans approved of same-sex marriage.

Contrast that with the history of support for interracial marriage. In 1948, when California’s Supreme Court struck down a law forbidding interracial marriage, 90 percent of Americas opposed it. 

By 1967, when the U.S. Supreme Court legalized interracial marriage nationally, some 72 percent were still opposed. Opponents only became a minority 43 years after the California decision and 24 years after the Supreme Court ruling. Today, 86 percent approve. 

Despite pushes from courts and legislatures, support for interracial marriage increased by about 1.2 points a year, a slower rate than acceptance of same-sex marriage.

That brings us to President Obama’s historic embrace of marriage equality and the poll question I hate. Posed about a variety of issues, in this case the offending item asked whether “President Obama’s support of same-sex marriage make[s] you more likely to vote for him, less likely to vote for him or doesn’t … make any difference.” Though 60 percent admitted it made no difference, by a 13-point margin people said they were less, rather than more, likely to vote for him as a result of his statement. 

Gallup also reported 53 percent support for gay marriage, a gigantic 9-point jump in just one year. 

If intensity is on the negative side, it’s certainly possible for folks to favor gay marriage in the net, while being more likely to oppose the president as a result of his view. However, most evidence suggests intensity is not unique to opponents. A Pew poll found equal numbers strongly favoring and strongly opposing gay marriage.

A deeper dive into the Gallup data reveals that about half of those ostensibly less likely to support the president are Republicans. With Romney consistently garnering about 90 percent of Republicans, what does it mean to say that half of them are “less likely” to vote for Obama because of his gay-marriage stance? Less likely than “not at all”? They weren’t going to vote for the president no matter what he said about same-sex marriage.

Independents break about evenly in Gallup’s horserace, and 23 percent report being “less” likely to vote for Obama. I’d guess that over 80 percent of that 23 percent come from the ranks of those already opposing Obama. So will 23 percent of independents defect on this issue? Not likely at all. 

With Gallup tracking daily, we can ask the ultimate question — did the decision hurt the president? Going into his announcement, Obama enjoyed a 3-point lead; today it’s 1. That 2-point difference could be caused by a range of factors, from sampling error to the economy to other news. Maybe even a touch of gay marriage, but nowhere near the level of defection predicted by the more/less-likely-to-vote question, once again casting grave doubt on its utility.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the majority leader of the Senate and the Democratic whip in the House.