September 29, 2006
Researchers have long known it’s difficult to count and classify men and women on the basis of their sexual orientation. Because homosexuality is stigmatized, many people are reluctant to disclose that they’re attracted to others of their same sex, or that they’ve had homosexual sex.
Two recently published studies offer interesting data relevant to this problem.
In a study conducted by Preeti Pathela and colleagues (reported in the Annals of Internal Medicine) nearly 4200 New York City men were interviewed by telephone and asked 130 questions about health-related matters. Embedded in the demographic questions midway through the survey was a question about the man’s sexual orientation. Later, at the end of the survey, each man was asked about the number of men and women with whom he’d had sex during the previous 12 months.
Of the men who labeled their sexual orientation and reported having sex in the past year:
- 85.8% identified as straight and reported sex only with women
- 3.3% identified as gay and reported sex only with men
- 1.1% identified as bisexual and reported sex with men, women, or both.
- 8.9% identified as straight and reported sex only with men
- 0.7% identified as straight and reported sex with women and men.
Combining the last two groups, nearly 10% of the men identified themselves as straight but had at least one male sexual partner in the previous 12 months. About 70% of these men were married. Nearly all reported having sex with only one partner in the past year.
In their published report Dr. Pathela et al. acknowledge various ways in which the phrasing or ordering of the questions might have affected the results. Nevertheless, their data underscore the fact that the labels people use for their sexual orientation, such as gay and straight, don’t always fit neatly with their actual sexual behavior. A significant minority of self-identified straight men in New York engage in sex with other men.
Based on another recently published paper, the New York data might actually understate the case.
In Public Opinion Quarterly, Maria Villarroel, Charles Turner, and their colleagues report data from a large-scale telephone survey conducted with a national sample and a Baltimore (MD) sample. The participants (age 18-45 years) were asked questions about their sexual behavior and attractions.
The participants were randomly assigned to one of two groups. Either they were interviewed by a “live” person or they completed most of the survey using a computer-automated system in which they responded to questions by using the buttons on their touch-tone telephone. The researchers expected people to be more candid in reporting about their sexuality when they didn’t have to speak to a live person.
Sure enough, those who were “interviewed” by the computer were significantly more likely to say they were attracted to people of their same sex (17.8% vs 12.8% of those interviewed by a live person). They also were more likely to report having sex with a person of their same sex (14.2% vs 9.1%).
Follow-up analyses revealed that the computerized interview was especially likely to get more reports of homosexual behavior in less tolerant locales — outside large cities and outside the Northeast and Pacific Coast. The computer also elicited more reports of same-sex behavior from people who might be particularly cautious about revealing such information — those who were currently married and had children in their home.
Together, these studies underscore the difficulties inherent in obtaining reliable data about stigmatized sexual behavior. Many people are reluctant to disclose information about their same-sex attractions and behavior, and many men (and probably women too) who call themselves straight have nevertheless had same-sex sexual partners.
For the full reports, see:
- Pathela et al. (2006). Discordance Between Sexual Behavior and Self-Reported Sexual Identity: A Population-Based Survey of New York City Men. Annals of Internal Medicine, v. 145, pp. 416-425 .
- Villarroel et al. (2006). Same-Gender Sex in the United States: Impact of T-ACASI on Prevalence Etimates. Public Opinion Quarterly, v. 70, pp. 166-196.