Gender differences in schizophrenia are of great interest to researchers, and some have recently concluded that female patients suffer from a more benign form of the illness. However, the research findings do not support this conclusion consistently, and some reports suggest greater impairment in female patients. In this article, we discuss the potential effects of sampling biases on the findings from studies that compare male and female patients. More specifically, we assume that females do manifest a less severe schizophrenic illness than males, and we propose that sex differences in severity thresholds for voluntary and involuntary treatment are contributing to inconsistencies in the research findings. Some other sources of sampling bias that may influence findings on gender differences are also discussed.