Lesbians and their heterosexual sisters were compared on demographic variab
les and mental health subscales, so that the feasibility of using heterosex
ual sisters as a control group for lesbians could be investigated. Lesbians
were significantly more educated more likely to live in urban areas, and m
ore geographically mobile than their heterosexual sisters. Heterosexual sis
ters were more likely than lesbians to be married and homemakers, to have c
hildren, and to identify with a formal religion. There was no difference in
mental health, but lesbians had higher self-esteem. When all respondents w
ere included bisexual women had significantly poorer mental health than did
lesbians and heterosexual women. This is the first study to use sisters as
a control group in lesbian research.