Five sets of criteria for the diagnosis of schizophrenia were compared on 47 hospitalized patients who fit New York Research Diagnostic Criteria (RDC) for chronic schizophrenia. The other diagnostic systems used were the Flexible System. Feighner Criteria, New Haven Schizophrenia Index (NHSI), and Taylor-Abrams (TA) Criteria. The NSHI, the broadest system examined, fit all of the patients studied. In contrast, only 30 patients (63% of the sample) were diagnosed schizophrenic using TA Criteria. When the Strauss-Carpenter prognostic scale was used, patients who fit TA Criteria for schizophrenia did not differ in the prognostic scores from the remainder of the sample. In addition, the two groups did not differ in the familial rate of schizophrenia spectrum disorders. The usefulness of narrow versus broad diagnostic criteria in schizophrenia research is discussed.