A program to teach behavioral analysis principles and applications to parents and other caretakers of disturbing children (normal, retarded, neurotic, psychotic, and others) was developed while serving 50 families. A nine-family study was then done comparing three 12-week treatments (minimum contact 1, nonoperant 1, and operant 1) in terms of therapeutic process and outcome. A second 12-week treatment period 3 weeks after the first was designed, in part, to offer operant 2 to the six families who had not received operant in treatment period 1. Three measurement periods occurred: one before treatment 1, one after treatment 1, and another after treatment 2. Process data were collected during all treatment periods; and outcome data, based on objective ratings of videotapes of parent-child interactions and parents' psychological test performances were collected during all measurement periods. Children in all groups improved in general behavior categories and in specific-to-each-family categories. Operant groups seemed better but there were not enough cases for statistical significance. The psychological tests of the six operant-only parent pairs improved more than did the three nonoperant-then-operant parent pairs. Recommendations for further research and applications are presented. (Author/JD)