The authors assessed the reliability and validity of the DSM-III criteria for schizotypal personality disorder (SPD) and their measurement by the Schedule for Schizotypal Personalities (SSP) developed by Baron. Additional criteria not in DSM-III were also explored. Consensual clinical ratings of videotaped interviews of subjects were used to validate item content of the SSP and the schizotypal diagnosis. The SSP assessments, with the exception of odd speech, were found to be reliable and valid. Higher cutoff scores than those recommended by Baron may be helpful in discriminating schizotypal from other psychiatric patients. Six of eight DSM-III criteria and all experimental criteria were highly specific for SPD cases but varied widely in their sensitivities. Based on these pilot data, the authors recommend combining all of the criteria into four conceptual categories: 1) self-report cognitive-perceptual disturbances, 2) observable disorders of thought and communication, 3) deficits in drive or affect, and 4) interpersonal difficulties. When three of these were required to make the SPD diagnosis, all of the subjects in the present data set were correctly identified.