June 9, 2008
USA Today Poll: What’s Wrong With This Picture?
From the June 3 USA Today:
Six in 10 Americans say the government should not regulate whether gays and lesbians can marry the people they choose, a survey finds…. The USA Today/Gallup Poll found that 63% of adults say same-sex marriage is “strictly a private decision” between two people. That the government has the right “to prohibit or allow” such marriages was stated by 33%, and 4% had no opinion.
The article goes on to report that “a majority of respondents at every level of education and income say same-sex marriage is ‘strictly private.'” This was true in every geographic region, among all age groups under 65 years, and among people who say a relative, friend or co-worker personally has told them he or she was gay or lesbian.
Sixty-three percent? Has a sea change occurred in American public opinion, with a clear majority now supporting marriage equality?
Probably not.
The Results in Context
When we compare the USA Today findings with those of other respected national polls, the inconsistencies are glaring. For example, surveys conducted by the Pew Forum on Religion & Public Life have found that:
In the time since the Massachusetts high court declared the state’s ban on gay marriage unconstitutional in 2003, public opinion on the issue has remained relatively stable. Indeed, majorities of Americans have consistently opposed legalizing same-sex marriage — from 53% opposed in a summer 2003 survey conducted by [Pew], to 55% opposed in an August 2007 Pew survey. The 2007 poll found 36% of the public in favor of allowing gay and lesbian couples to wed, about the same as in 2003.
A May 11th survey conducted by Gallup (this one without USA Today as a sponsor) asked respondents “Do you think marriages between same-sex couples should or should not be recognized by the law as valid, with the same rights as traditional marriages?” 56% said they should not be recognized by the law as valid, while 40% said they should be valid. Like the Pew Center’s surveys, Gallup has found these opinions to be fairly stable during recent years. In 2004, for example, 55% said same-sex marriages should not be valid, and 42% said they should be valid.
Thus, data from the Pew Center and Gallup (and others as well) don’t support the conclusion that a substantial majority of Americans oppose government prohibitions against marriages between two people of the same sex.
Reconsidering the USA Today Data
How can we explain the anomalous USA Today findings? When a survey’s results are so at odds with other polls, it’s a good idea to scrutinize its methodology even more closely than usual.
Let’s assume that the USA Today sample wasn’t dramatically less representative of the population than those used by the earlier Gallup surveys, and instead focus on how the question’s wording and its location in the interview might have affected the outcome.
Unfortunately, the USA Today article doesn’t clearly detail the question wording. Apparently, however, survey participants were presented with pairs of statements about different types of marriages. From each pair, they selected the one that better matched their own opinion. The order of questions seems to have been the same for everyone.
The questions went something like this (I’ve highlighted key differences):
1. “In marriage involving two people of different religions, the decision to marry should be strictly a private decision between the two people who want to marry” [OR] “The government has the right to pass laws to prohibit or allow such marriages.”
2. “In marriage involving two people of different races, the decision to marry should be strictly a private decision between the two people who want to marry” [OR] “The government has the right to pass laws to prohibit or allow such marriages.”
3. “In marriage involving two people of the same sex, the decision to marry should be strictly a private decision between the two people who want to marry” [OR] “The government has the right to pass laws to prohibit or allow such marriages.”
The choice is always between saying that the decision to marry is a private matter versus endorsing the view that the government has the right to pass laws about marriage. The problem, I suspect, is that these alternatives aren’t mutually exclusive in the minds of most people.
Deciding to marry is not the same thing as actually having that marriage recognized by the state. Opponents of marriage equality might agree that a decision by two women or two men to marry is a private one, even as they oppose State recognition of that marriage. Their likelihood of endorsing the “private matter” option might have been increased by their responses to the preceding questions about interfaith and interracial marriages, which could have made salient their belief that adults’ marital decisions are private.
USA Today probably included the phrase “strictly a private decision” to clearly distinguish that option from the “pass laws” alternative. But it apparently didn’t work for many people.
To be sure, more respondents chose the “private matter” option for interfaith and interracial marriages (97% and 95%, respectively) than for same-sex relationships (63%). So respondents weren’t automatically selecting the “private matter” option for every question. Perhaps the first two questions were easier to answer — respondents knew that they considered those types of marriage decisions to be private and that the government has no legal right to prohibit them. Most respondents probably felt that same-sex coupling is also a private decision but many also believed the government can refuse to recognize them. Faced with the question’s ambiguities (e.g., was it asking whether the government could prevent the decision to marry or the marriage itself?), a large number of respondents selected the privacy option.
In two subsequent questions, relatively few said that polygamous marriages or marriages between people under 16 are strictly private (30% and 18%, respectively). This probably reflects a view among many respondents that such marriages aren’t simply a private decision (e.g., that young minors aren’t capable of making a decision to marry) and that the State has a right to prohibit them. Such reactions were likely reinforced by news coverage during recent months of the polygamous Fundamentalist Church of Jesus Christ of Latter Day Saints, which included allegations of forced marriages and physical abuse of young girls.
Whatever the reason, the USA Today poll results just don’t fit with what we know about current opinion on marriage equality.
The lessons: Be a critical consumer of empirical research. Always read the wording of a survey’s questions and, whenever possible, compare the findings to other available data.