The responses were coded according to one or more of thirteen themes by independent raters blind to the subjects' gender. Six of the themes were identified as "feminine" themes and seven as "masculine" themes on the basis of previous work by Gilligan 1982, Kohlberg 1976, Likona and others. Only one association between gender and the presence of any given theme reached statistical significance; there is little evidence to support the idea that men and women differ in their reports of how they think about moral dilemmas. For all subjects, the average proportion of possible feminine themes in response was higher than the proportion of possible masculine themes. This finding supports the idea that an exclusive focus on such themes such as rights and responsibilities will fail to capture many of the considerations that subjects regard as most important. 2 tables, 25 references. (Author abstract modified)