Depression has been one of the last categories of psychopathology to be seriously investigated. A self-control behavior therapy program for depression was evaluated against a therapist-control behavior therapy program in an attempt to isolate the effects of control on alleviation of depression and to improve research designs. Clinically depressed obese subjects (N=22) were randomly assigned to either a self-control or therapist-control behavior therapy program that met for six weekly treatment sessions and three monthly maintenance sessions. There was a significant treatment group and assessment period interaction on the Depression Scale of the Minnesota Multiphasic Personality Inventory, with the therapist-control group significantly more improved at posttest and follow-up. In a second experiment, 33 clinically depressed subjects were matched to one of three groups: self-control, therapist-control, or attention-placebo. There was a significant treatment group and assessment period interaction on the Lewinsohn Pleasant Events Schedule Reinforcement Potential, with therapist-control subjects significantly higher than self-control or attention-placebo subjects. Friends and relatives considered therapist-control subjects to have changed significantly more than self-control or attention-placebo subjects. Findings demonstrate that self-report, overt behavioral measures, and peer ratings can be used to assess depression. (Author)