One of the more difficult issues in the development of each edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM) has been a possible sex bias in the personality disorder diagnoses. A substantial amount of discussion and research on this issue has occurred since the 1980 publication of the third edition of the DSM. It is now apparent that there are a number of different ways in which the differential sex prevalence rates for the DSM-IV personality disorders could reflect a sex bias, including biased diagnostic constructs, biased thresholds for diagnosis, biased population sampling, biased application of diagnostic criteria, biased instruments of assessment, and biased diagnostic criteria. It is important to understand these different forms of sex bias, as each can occur independently of, be confused with, and interact with one another. The purpose of this paper is to differentiate among, and to describe the support for, each of these different forms of sex bias, with the hope of contributing to their recognition and ultimate resolution.