Advancing Sexual and Reproductive Health and Rights
Perspectives on Sexual and Reproductive Health
Volume 40, Number 4, December 2008

Teenagers' Sexual Identity May Not Reflect Behavior; Both Are Linked to Risk

Teenage women’s sexual identity frequently does not mesh with their sexual behavior, and as analyses of data from the Massachusetts Youth Risk Behavior Survey show, both identity and behavior are independently related to the likelihood of risky outcomes among young women.1 Participants who considered themselves lesbian or bisexual, as well as those who were unsure of their sexual identity, were more likely than self-identified heterosexuals to say they had ever been coerced to have sex; women who had had partners of both sexes were more likely than those who had had sex only with males to have experienced coercion, but those with only female partners had reduced odds of this outcome. Lesbians and bisexual women reporting any male partners had an elevated likelihood of having been pregnant, and participants reporting sexual experience with women were more likely than others to have had four or more partners.

The analyses were based on data from four rounds (1995, 1997, 1999 and 2001) of the cross-sectional survey, which is conducted among public high school students throughout the state, and included only sexually experienced participants. Analysts used chi-square tests to examine demographic and risk-related characteristics of subgroups defined by sexual identity and sex of partners, and logistic regression to examine associations between these dimensions of sexual orientation and selected outcomes related to HIV risk.

A total of 3,973 students were included in the analytic sample. Of these, 3,666 identified themselves as heterosexual, 21 as lesbian and 163 as bisexual; 113 were not sure of their sexual identity, and 10 did not answer the question. (Because of small numbers, the researchers combined all women identifying as sexual minorities—i.e., lesbians and bisexuals—for analysis.) Some 3,714 of the women said they had had sex only with men, 79 only with women and 180 with both. Nearly all women who reported only male partners considered themselves heterosexual, but 4% gave other responses. By contrast, among those who had had sex exclusively with females, only 14% identified as lesbian or bisexual; 82% thought of themselves as heterosexual, and 4% were unsure. Fifty-eight percent of those reporting sexual experience with both men and women said that they were lesbian or bisexual; 31% identified as heterosexual, and the rest were unsure.

Participants were about 16 years old, on average, and mean age did not differ by sexual identity or sex of partners. Among identity subgroups, women who were unsure were the least likely to be white and the most likely to be Hispanic; heterosexuals were the least likely to have immigrated in the last six years. Comparisons by partners’ sex showed that participants with exclusively same-sex experience were the least likely to be white and the most likely to be recent immigrants.

At the bivariate level, a wide range of risk-related outcomes differed by sexual identity. Women who identified as sexual minorities were the most likely to report injection-drug use, very early intercourse, multiple lifetime or recent partners, pregnancy or STD history, and sexual coercion; they were the least likely to report having received AIDS education in school. Those who were unsure about their identity reported the highest incidence of dating violence and, if they had had male partners, the lowest level of condom use at last sex. Heterosexuals had the lowest risk profile.

All of these outcomes except for condom use at last sex also differed by sexual behavior. Participants who had had partners of both sexes exhibited the highest level of risk; those reporting only male partners generally were at lowest risk.

An initial set of multivariate analyses examined predictors of sexual coercion and receipt of AIDS education. The findings indicate a positive association between minority sexual identity or uncertainty and reports of sexual coercion (odds ratios, 1.7 and 1.6, respectively, in a model controlling for age, ethnicity, immigrant status, survey year and partners’ sex), but no association between identity and AIDS education. Sexual coercion was negatively associated with having had only female partners and positively associated with having had partners of both sexes (0.5 and 2.1, respectively, in a model including sexual identity). Receipt of AIDS education was negatively associated with reporting only female partners or partners of both sexes (0.4–0.5).

Analyses that included sexual coercion and AIDS education among the controls showed few relationships between sexual identity and the risk-related outcomes. Lesbian and bisexual women who had ever had a male partner were more likely than heterosexuals to have been pregnant (odds ratio, 2.2), and women who were unsure of their sexual identity had reduced odds of reporting condom use at last sex if they had had only male partners (0.5). Partners’ sex was associated with three outcomes. Women reporting only female partners or partners of both sexes had elevated odds of reporting injection-drug use (3.2 and 5.0, respectively) and at least four partners (2.3 and 3.5); in addition, those who had had sex with both women and men had elevated odds of reporting two or more recent partners (2.6).

Sexual coercion was the most consistent predictor of risk outcomes. The likelihood of reporting condom use at last sex was reduced among women who had experienced coercion (odds ratio, 0.7), and the likelihood of reporting each of the other outcomes was elevated (1.9–3.3). Receipt of AIDS education was negatively associated with injection-drug use, experience with multiple lifetime and recent partners, and STD diagnoses (0.3–0.5).

While acknowledging several study limitations (including the grouping together of all sexual minorities, the age of the data and uncertainty about participants’ understanding of some questions), the analysts say that their findings show that adolescents whose sexual identities or behaviors “place them outside the heterosexual majority” are at increased risk of unhealthy outcomes. They therefore stress the need for interventions that “are sensitive to the complexity of sexual-orientation development during adolescence and that are effective in helping young people make healthy choices.”


1. Goodenow C et al, Dimensions of sexual orientation and HIV-related risk among adolescent females: evidence from a statewide survey, American Journal of Public Health, 2008, 98(6):1051–1058.