Author discusses from a technical/statistical point of view the problems associated with validation of tests. He presents information on criterion problems, global approach for projectives, and higher order interaction effects among test variables. A key factor in any test interpretation is the clinical skill of the interpretor. (JLH)