Expert Commentary

Same-sex unions and gay marriage: Revisiting anti-gay sentiment and population figures

2013 paper for the National Bureau of Economic Research using two survey methods to measure attitudes toward (and by) lesbian, gay, bisexual and transgender individuals in the United States.

On Oct. 6, 2014, the U.S. Supreme Court let stand rulings at three Federal appeals courts, clearing the way for same-sex marriages in Indiana, Oklahoma, Utah, Virginia and Wisconsin, bringing the total of U.S. states permitting such unions to 24, in addition to the District of Columbia.

Although dynamics could change if the Court eventually decides to rule on the matter, its latest action could lead to even more states permiting gay marriage: The three appeals courts in question have jurisdiction over six other states that currently ban same-sex unions — Colorado, Kansas, North and South Carolina, West Virginia and Wyoming.

In the months following the June 2013 Supreme Court ruling that struck down key parts of the 1996 Defense of Marriage Act, states have continued to take measures to legalize same-sex marriage, most recently Illinois and Hawaii. New Jersey also began to recognize them after Governor Chris Christie withdrew a challenge to such unions, and in November federal judges struck down gay-marriage bans in Arkansas and Mississippi. While many continue to oppose gay marriage, recent polls indicate that attitudes are softening even among conservatives. However, some polling indicates that a majority would like to see the issue resolved by individual states — a February 2014 New York Times/CBS News poll indicated that nearly two-thirds of Americans would prefer a state-based approach.

On the international stage, Pope Francis has made public comments on not judging homosexuality, setting a different tone for the Catholic Church. In Russia, however, President Vladimir Putin has signed a law banning “propaganda on nontraditional sexual relationships,” seen as part of a larger crackdown on advocates’ efforts to improve gay rights in the country. The law and its heavy-handed enforcement cast a pall over the 2014 Winter Olympics, in the Black Sea town of Sochi.

A June 2013 Pew survey of lesbian, gay, bisexual and transgender adults found that the vast majority (92%) felt that U.S. society had become more accepting over the past decade. At the same time, discrimination was common: Nearly 40% said they’d experienced rejection because of their sexual orientation or gender identity; 30% had received threats or been physically attacked; and 58% said they’ve been the target of slurs or jokes.

A 2013 paper for the National Bureau of Economic Research, “The Size of the LGBT Population and the Magnitude of Anti-Gay Sentiment are Substantially Underestimated,” looks at the possibility that discrimination may be more pervasive than previously understood — and that the extent of the U.S. gay population remains under-counted. The researchers — Katherine B. Coffman and Lucas C. Coffman of Ohio State University and Keith M. Marzilli Ericson of Boston University — suggest that survey respondents are often hesitant to express negative attitudes toward the LGBT community if they believe they will sound intolerant. Previous research indicates that honest replies increase when survey participants are guaranteed anonymity and privacy, and the use of computers has been shown to be effective. But is this sufficient for eliciting completely truthful responses to deeply personal questions?

The researchers used an experiment design that allowed them to compare results from a traditional anonymous survey with one that uses an item count technique (ICT), also called an “unmatched count technique.” Rather than asking participants to answer sensitive questions directly, ICT surveys typically involve asking a control group to look at a list of neutral, non-sensitive items and report how many are true for them. A second group is given the same list, plus one sensitive, personal item, such as an indication of sexual orientation. The researcher is then able to estimate the population mean for the sensitive item by comparing it with the control group. (This technique has also been used to measure subjects such as risky driving by veterans and voter turnout.)

In the NBER study, which was conducted in November 2012, participants from Amazon’s Mechanical Turk service were randomly given “direct report” or “veiled report” treatments. Each group was presented with both neutral statements and sexuality-related questions. The direct-report group was asked to reply to sensitive questions anonymously and privately on a computer, imitating the design of many current surveys. The veiled-report group was given the same neutral and sensitive statements, but merely had to report the number of statements — innocuous or sensitive — that applied to them. (Note that percentage points and percent are not synonymous; our article on statistics for journalists has more information.)

Key results from the study include:

  • The veiled-report treatment significantly raised the overall number of answers to sensitive questions: Reports of non-heterosexual identity rose by 65%, same-sex sexual experiences by 59% and directionally same-sex attraction by 9.4%.
  • The veiled report increased measured rates of anti-gay sentiment: Participants were 67% more likely to express disapproval of an openly gay manager at work, 71% more likely to say it should be legal to discriminate in hiring on the basis of sexual orientation, 22% less likely to support the legality of same-sex marriage, 46% less likely to support adoption by same-sex couples, and 32% less likely to state they believe homosexuality is a choice.
  • “Individuals in demographic categories that other research has identified as more openly anti-gay — Christians, African-Americans, and older populations — are more likely to lie about their sexual identity without a veil.” For example, Christians increased reports of non-heterosexuality by 13 percentage points and same-sex sexual experiences by 14 percentage points in the veiled report as opposed to the direct report.

The authors conclude that “standard methods of eliciting respondents’ sexual orientation and behavior underestimate the true fraction of individuals who do not identify as heterosexual and who have had a same-sex sexual encounter.” They argue that such findings — although not from a perfectly representative sample of the U.S. population — suggest “a societal stigma of being LGBT,” coupled with the fact that “individuals are reluctant to report that they have attitudes or policy opinions that are not accepting of LGBT individuals, consistent with a stigma of holding anti-gay sentiments.”

The Pew Research Center also analyzes the implications of this study and provides more detail on the “veiled report” survey techniques that were employed.

Keywords: gay and lesbian issues, equality, bisexuality, discrimination

About The Author