DSM-II-R diagnostic criteria seem unsuited to the task faced by clinicians dealing with sex offenders in that it allows only a small proportion of sex offenders to be identified as having a disorder. The areas that require assessment in clinics are cognitive factors such as denial and minimization regarding the offense; and the social competence, lifestyle, and sexual behavior of the sex offender. In the evaluation of sexual preference, there are many unresolved issues concerning the measurement of erectile responses. The reliability of these measures has not been as clearly established as it should be, the degree to which they are vulnerable to deliberate faking by the respondent needs further analysis, and the best way to represent the resultant data has not been resolved. Treatment for sex offenders is effective; cognitive-behavioral programs seem to offer the best hope with antiandrogens having a valuable adjunctive role for some patients. 1 table and 115 references (Author abstract modified)