In two studies, 264 male and female managers reviewed information about the job performance of a person portrayed as either a man or a woman and, if a woman, as either an affirmative action hire or not. As expected, subjects rated female affirmative action hires as less competent and recommended smaller salary increases for them than for men and women not associated with affirmative action. This pattern held even when disconfirming performance information was provided if that information was ambiguous either with regard to degree of success (Study 1) or with regard to who was responsible for the success (Study 2).