This question comes up every so often, so I thought I'd pass along what seems to be the best data out there -- from Laumann et al., The Social Organization of Sexuality 311 (1994). All numbers are percentages.
1. Sexual partners:
|Last year (men / women)||Past 5 years (men / women)||Since age 18 (men / women)||Since puberty (men / women)|
|No partners||10.5 / 13.3||5.9 / 7.1||3.8 / 3.4||3.3 / 2.2|
|Opposite gender only||86.8 / 85.4||90.0 / 90.7||91.3 / 92.5||90.3 / 94.3|
|Both men and women||0.7 / 0.3||2.1 / 1.4||4.0 / 3.7||5.8 / 3.3|
|Same gender ony||2.0 / 1.0||2.0 / 0.8||0.9 / 0.4||0.6 / 0.2|
2. Sexual identity ("Do you think of yourself as heterosexual, homosexual, bisexual, or something else?"):
3. Sexual attraction ("In general are you sexually attracted to only men, mostly men, both men and women, mostly women, only women?"):
|Only opposite gender||93.8||95.6|
|Mostly opposite gender||2.6||2.7|
|Mostly same gender||0.7||0.6|
|Only same gender||2.4||0.3|
Naturally one has to be cautious about even well-conducted random studies of small sexual minorities, especially when some respondents might lie. Also, note that even though the study tried to be precise in the questions it asked, other studies might not, or might focus on different questions -- whether someone is "gay" or "lesbian" is not unambiguously defined, and the definitions may vary from survey to survey and respondent to respondent. Still, this seems to be the best approximation I've seen.
Related Posts (on one page):