We used a new self-report measure, the Kings Schizotypy Questionnaire (KSQ; Williams, M. The psychometric assessment of schizotypal personality. PhD thesis. Institute of Psychiatry, University of London, 1993), to investigate schizotypy as a quantitative measure of familial liability to schizophrenia. The KSQ was administered to 135 DSM-IV schizophrenia probands, 153 of their healthy first-degree relatives, and 267 control subjects. We found that the questionnaire clearly differentiated schizophrenic from non-schizophrenic individuals, but failed to differentiate the relatives from controls. Possible reasons for this include defensive responding among relatives, self-selection bias among relatives, differences in data collection methods, and the possibility that positive aspects of schizotypy may not be closely related to familial liability to schizophrenia.