Correctional authorities must evaluate all offenders in order to assign them to appropriate treatment, rehabilitative, or special education programs. Studies on traditional methods of assessment are reviewed, including clinical diagnostic methods, the Minnesota Multiphasic Personality Inventory, and the industrial placement and selection model used by the Kentucky correctional system. Genetic research on XYY genes is cited as an example of the biosocial approach. Because intelligence quotients (IQ'S) are a major criteria for certifying mental retardation of the juvenile offender, several research reports on IQ tests are summarized. Both criterion and norm referenced measurements are examined. Situational variables which can affect test scores are discussed, such as educational background, anxiety, motivation, characteristics of the examiner, and cultural bias. Authorities agree that a culture-fair test does not exist at present, and many doubt that any test can be equally fair to more than a single cultural group. Minorities have pressured for culturally adjusted IQ tests, greater representation in test selection, and more control over situational factors. A different approach to minority testing is described which advocates a multicriteria approach using diagnosis of organic disfunction, assessment of social role performance in a variety of social systems, and identification of learning skill deficiencies that may interfere with academic skill development. A bibliography provides 13 references.