Group homes are community-based residences for small groups of young people who cannot live with their own families. The first small group residential treatment program for juvenile delinquents in such a home was the Highfields project, developed in New Jersey in 1952. By the mid-1960's, the incarceration of delinquents and status offenders was recognized as an ineffective solution, and by 1978 more than 1,000 group homes in about 37 States had been established. Compared to day-care treatment centers and institutions, the homes rank first in a number or respects, including the encouragement of respect among racial groups, the provision of clear expectations and good staff role models, more involvement in the community, and better preparation for release. An examination of evaluations covering 1,000 homes yielded a number of hypotheses which deserve further study. These hypotheses include the following possibilities: residents who live longest in the homes have lower recidivism rates than those with shorter stays, those who complete their programs successfully seem to have lower recidivism rates, girls seem to adjust to the programs better than boys and have lower recidivism rates after 'graduation,' self-concept seems to be a major factor in group home success, and a population of five or six residents seems to be most successful. Although Great Britain has given little attention to group homes, a new project, Special Family Placement, in Kent, is in many ways similar to the American group homes. However, the English home has two advantages. The voluntary character of the placement is made much more explicit, since the individual's will to change is very important, and the home has not been pressured into adopting any preformed treatment techniques. Sixteen reference notes are provided.