There are few data sources on the prevalence of same-sex sexual orientation in England.We aimed to measure the prevalence of same-sex orientation and behavior in the English general population and assess the impact of enquiry format on reporting. The Adult Psychiatric Morbidity Survey 2007 used a multi-stage, stratified probability-sampling design (n=7,403). Two questions addressed sexual orientation and sexual partnership and each had two versions. Version A of the sexual orientation question used "homosexual." Version B used "gay or lesbian." Version A of the sexual partnership question required participants who had male and female partners to say which was predominant, while Version B had a midpoint response option: "about equally with men and women." Participants were randomized between versions. Overall, 5.3% of men and 5.6%of women reported they were not entirely heterosexual. The question using "gay or lesbian" elicited higher (though not statistically significant) reporting of non-heterosexual orientation than the question using "homosexual." A significantly larger proportion of men and women (96.0 and 96.1%) reported entirely heterosexual partnerships in response to Version A of the partnership question than in response to Version B (94.0 and 92.9%) where Version B asked specifically about "kissing, touching, intercourse, or any other form of sex." These figures constitute the first national prevalence data on combined sexual orientation and sexual behavior in England, based on a random probability sample of the general population. They demonstrate that people are willing to report their sexual orientation in survey research, but reporting is sensitive to question wording.