GROUP HOMES ARE DEFINED AS COMMUNITY-BASED SUPERVISED RESIDENCES FOR SMALL GROUPS OF CHILDREN WHO ARE UNABLE TO LIVE WITH THEIR OWN FAMILIES. GROUP HOMES FIRST DEVELOPED IN RESPONSE TO THE NEEDS OF CHILDREN DURING WORLD WAR II. THE FIRST ATTEMPT AT A COMMUNITY-BASED GROUP HOME FOR JUVENILE DELINQUENTS IN THE U.S.A. WAS IN NEW JERSEY IN 1952. ACHIEVEMENT PLACE, A BEHAVIOR MODIFICATION GROUP HOME IN KANSAS, WAS INITIATED IN THE 1960'S AND BECAME A MODEL FOR MANY OTHER PROGRAMS. SINCE THEN THE MOVEMENT HAS FLOURISHED, AND GROUP HOMES NOW NUMBER OVER 1,000 AND ARE FOUND IN AT LEAST 39 STATES. FACTORS LEADING TO THE GROWTH OF THE MOVEMENT INCLUDED THE EMPHASIS ON DEINSTITUTIONALIZATION, COST-EFFICIENCY, OVERCROWDING IN OTHER INSTITUTIONS, A GROWING CONCERN FOR DIVERSIFIED TREATMENT, LEGISLATION, AND FEDERAL FUNDING. BASED ON A STUDY OF EVALUATION PAPERS REFLECTING DATA ON 140 GROUP HOMES FOR JUVENILE DELINQUENTS IN THE 4 MAJOR REGIONS OF AMERICA, A GROUP HOME PROFILE EMERGES. MOST HAVE 5 OR FEWER RESIDENTS, ALTHOUGH SOME HOMES HAD OVER 10 RESIDENTS. RESIDENTS ARE GENERALLY UNDER 18 YEARS OLD. THE AVERAGE LENGTH OF STAY IS 2 TO 3 MONTHS; A SIGNIFICANT CORRELATION WAS FOUND BETWEEN LENGTH OF STAY AND SUCCESS OF THE PROGRAM. ANNUAL COST PER CHILD RANGED FROM ABOUT $3,000 UP TO ALMOST $25,000. IN GENERAL, GROUP HOMES ARE LESS EXPENSIVE THAN INSTITUTIONAL CARE. ADMISSION CRITERIA VARY WIDELY, AS DO TREATMENT APPROACHES. THE MOST FREQUENTLY USED TREATMENT APPROACHES INCLUDE MEDICAL, BEHAVIOR MODIFICATION, EDUCATION, USE OF COMMUNITY RESOURCES, AND GROUP THERAPY. MOST HOMES LIST THE HOME ENVIRONMENT AS A THERAPEUTIC APPROACH. THE USE OF THESE APPROACHES IN A NUMBER OF SPECIFIC GROUP HOME PROJECTS IS EXAMINED AND THE REASONS FOR SUCCESS OR LACK OF SUCCESS ARE DISCUSSED. MOST HOMES HAVE SIMPLE ORGANIZATIONAL SCHEMES, ALTHOUGH THERE IS A TENDENCY FOR INCREASED BUREAUCRATIZATION IN HOMES THE LONGER THEY EXIST. ALTHOUGH MOST EVALUATIONS LACK SYSTEMATIC CONTROLS, HIGH RECIDIVISM RATES SUGGEST THAT GROUP HOMES MAY NOT BE MORE EFFECTIVE THAN OTHER APPROACHES TO JUVENILE DELINQUENCY. DELINQUENTS WHO WERE SUCCESSFUL WITHIN THE GROUP HOME TEND TO ALSO BE SUCCESSFUL WHEN RELEASED. MOST GROUP HOMES UTILIZE A HOUSE PARENT MODEL. OCCUPATION/EDUCATION OF HOUSE PARENTS IS EXTREMELY DIVERSE, AND THERE IS A RELATIVELY HIGH RATE OF TURNOVER AMONG LIVE-IN STAFF. COMMUNITY-BASED TREATMENT CAN BE EITHER IN THE RESIDENTS' COMMUNITY OR NOT. REGARDLESS OF THE PROGRAM, COMMUNITY ACCEPTANCE, SUPPORT, AND INVOLVEMENT SEEM CRUCIAL TO PROGRAM SUCCESS. A STRATEGY FOR FACILITATING COMMUNITY ACCEPTANCE IS OUTLINED. FROM THESE FINDINGS, 21 HYPOTHESES RELATING TO GROUP HOMES, RECIDIVISM, AND TREATMENT ARE FORMULATED FOR FUTURE RESEARCH. EXTENSIVE TABULAR DATA AND REFERENCES ARE PROVIDED.