UTILIZATION OF RESEARCH FINDINGS OFTEN REQUIRES SOCIAL CHANGE CONSISTING OF THE CREATION OF A SOCIAL INNOVATION AIMED AT PROBLEM SOLUTION, AND THE IMPLEMENTATION OF THE MODEL. FOR A SOCIAL CHANGE MECHANISM TO BE BENEFICIAL AND ADEQUATE, IT MUST RESULT FROM PROBLEM-ORIENTED RESEARCH AND OFFER USEFUL SOLUTIONS. THERE MUST BE A DEMOCRATIC PARTICIPATION IN THE PROGRAM, AND PROVISIONS MUST BE INCLUDED FOR SCIENTIFIC EVALUATION. SOCIAL PROGRAM INNOVATION AND EVALUATION SHOULD BE LINKED DIRECTLY WITH A DISSEMINATION EFFORT. THIS EFFORT INVOLVES FOUR STAGES OF DEVELOPMENT: (1) THE APPROACH, IN WHICH DIFFERENT TARGETS ARE CONTACTED AND THEN EVALUATED; (2) THE PERSUASIVE EFFORT, IN WHICH THE SELECTED TARGET GROUPS ARE INTRODUCED TO THE SOCIAL CHANGE THROUGH VARIOUS TACTICS SUCH AS THE WRITTEN WORD, DEMONSTRATIONS, WORKSHOPS, AND ECONOMIC REINFORCEMENTS; (3) THE ACTIVATION PROCESS, IN WHICH THE MODEL IS IMPLEMENTED; AND (4) THE DIFFUSING PROCESS, IN WHICH THE MODEL, BEING SUCCESSFUL IN ONE INSTANCE, IS TRANSFERRED TO OTHER SIMILAR SITUATIONS. A CASE STUDY INVOLVING PATIENTS RELEASED FROM A MENTAL HOSPITAL IS PRESENTED IN CONJUNCTION WITH THESE COMPONENTS OF RESEARCH UTILIZATION. THE STUDY WAS CONDUCTED TO DEVELOP TREATMENT WHICH WOULD RESULT IN A LOWER RETURN RATE FOR THE CHRONIC MENTAL PATIENTS. A QUASI-EXPERIMENT INDICATED THAT CERTAIN DEMOGRAPHIC CHARACTERISTICS, SUCH AS MARITAL STATUS, DIAGNOSIS, AND LEGAL COMPETENCE, WERE RELATED TO RETURN RATES. AN ACTUAL EXPERIMENT WAS THEN CONDUCTED TO ASCERTAIN IF A TYPE OF TREATMENT COULD BE DEVELOPED TO REDUCE THE RATES, AND 4 DIFFERENT TREATMENTS WERE COMPARED ON 120 OUTCOMES FOR 3 DIFFERENT PATIENT GROUPS. RESULTS INDICATED THAT THE PROBLEM WAS BEING APPROACHED FROM THE WRONG ANGLE, AND THAT THESE PERSONS PROBABLY NEEDED TO BELONG TO SOME SORT OF SOCIAL ORGANIZATION. A 5-YEAR GROUP OF EXPERIMENTS ON THIS CONCEPT RESULTED IN THE ESTABLISHMENT OF THE LODGE, A SELF-OPERATED SOCIETY ORGANIZED IN THE COMMUNITY FOR CHRONIC MENTAL PATIENTS. THE RETURN RATE FOR MEMBERS OF THE LODGE WAS SIGNIFICANTLY LOWER THAN IT WAS BEFORE THE ORGANIZATION WAS ESTABLISHED. THE TASK OF IMPLEMENTATION WAS THEN CARRIED OUT IN 255 HOSPITALS, ON THE FEDERAL, STATE, AND FINALLY THE LOCAL LEVEL. THE PROCESSES OF DISSEMINATION ARE DESCRIBED AS THEY WERE USED IN THE LODGE EXPERIMENT. SUGGESTIONS FOR FURTHER INCREASING THE POSSIBILITY OF IMPLEMENTATION OF A SOCIAL PROGRAM ARE LISTED. (DAG) INCREASING THE POSSIBILLITY OF IMPLEMENTATION OF A SOCIAL PROGRAM ARE LISTED. (DAG)