A modified version of the Kansas Marital Satisfaction Scale was administered to 239 women who belonged to 8 different faith communities in Topeka, Kansas in 2001. 12 members of a Metropolitan (gay and lesbian) church identified themselves as lesbians. Lesbian respondents reported significantly higher relationship satisfaction scores than heterosexual women (Cohen's d = 0.69). Controlling for number of children and social desirability reduced the regression coefficient for sexual orientation to a statistically nonsignificant level. Using a matched sample of 12 heterosexual women compared with the lesbians yielded a nonsignificant result (ES = 0.31). Nevertheless, although not statistically significant, comparisons between lesbians and heterosexual women continued to feature effect sizes that represented higher satisfaction for lesbians with no children or with only one child when compared to heterosexual women with no children or only one child. It is apparent that methodology made a difference in the results obtained in this comparative study of lesbian and heterosexual relationships. It was also observed, among members of churches other than the Metropolitan church, that relationship satisfaction was significantly lower (d = 0.22) among females than males, including among wives compared to husbands; significant linear and cubic relationships between a single-item measure of relationship social desirability and relationship satisfaction were also observed.