Men (N= 264) and women (N= 343) students were shown one of 4 videotape versions of a female stimulus person (SP) being interviewed. The SP portrayed a Competent or Incompetent individual who was Feminine or Masculine in her interests. The S s, who had previously been given an Attitudes Toward Women Scale (AWS, Spence & Helmreich, 1972) rated the SP on several characteristics, including likability. A 2 × 2 ANOVA indicated that women S s significantly preferred the Competent and the Masculine SPs. For men, the variables interacted significantly, the Masculine-Competent SP being liked most, the Masculine-Incompetent least, and the two Feminine SP s being intermediate and similar to each other. When S s' AWS scores (high, medium, or low) were included, complex interactions were found between AWS and experimental condition, especially in the male groups.