Critics of personality research have asserted that consistency of personality is in the eye of the beholder and does not reflect differences between rated targets. The relative importance of perceiver and target effects is assessed with self- and other-ratings on five personality tests. Three members of the same family, both parents and an adolescent, supplied self-ratings as well as ratings of the two other members. Perceiver and target effects were assessed with analysis of variance. The correlation between ratings of the same target by different perceivers was compared with that between ratings of different targets by the same perceiver. Both types of analysis showed target effects to be more important than effects of the perceiver. The differences between the present study and those using free descriptions instead of personality tests are discussed.