Two basic factors influence mutual ratings of social groups: in-group favoritism (related to the evaluative aspects of a rating) and the perception of shared reality (related to the descriptive aspects). In two studies, we examine the usefulness of Peabody's (1968) method of separating evaluative and descriptive aspects of rating in intergroup judgments. In Study 1, Latvian and Russian students made different evaluations of both groups, but the same groups agreed on the descriptive ratings. In Study 2, male and female psychology students rated each other from own, in-group, and out-group perspectives. The participants did not show any in-group favoritism in their own ratings, but they expected their fellow students to be in-group biased. The participants agreed on the descriptive ratings of both groups. The results demonstrate that shared reality influences intergroup ratings, despite differences in evaluations.