Studies on race bias, social class bias, and gender bias are reviewed. Topics include psychodiagnosis and rating level of adjustment, the description of personality traits and psychiatric symptoms, the prediction of behavior, and treatment planning. Replicated findings indude race bias in the dffarential diagnosis of schizophrenia and psychotic afkctive disorders, gender bias in the dfhrential diagnosis of histrionic and antisocial personality disorders, race bias and gender bias in the prediction of violence, and social class bias in the referral of dients to psychotherapy. Recommendations for decreasing bias include (a) being aware of when biases are likely to occur, (b) adhering to diagnostic criteria, and (c) using statistical prediction rules to predict behavior. Key wods: cognitive biases, judgment, personality assessment, prediction, psychodiagnosis. f a i n Psycho1 Sci Prac 4:99-120, 19971Misconceptions exist about research on race bias, social class bias, and gender bias. For example, according to a recent article in The A P A Monitor (Sleek, 1996), Practitioners tend to judge lower-class patients as more seriously disturbed than higher-class patients. They detect schizophrenia in black people more than in whites. And they offen judge women patients as more disturbed than male patients presenting identical problems. @. 30)