Eighty-five sponsors in a Veterans Administration family care program were interviewed for the purpose of ascertaining whether or not the social environment provided a viable alternative to institutional living for the chronic mentally ill. Also, the impact of various descriptive factors of the setting on the social characteristics in the home was estimated. A level of restrictiveness scale (areas in which the freedom of the residents is restricted) and a level of deviation from normative living scale (areas in which residents are excluded from family activities) were utilized to measure the social characteristics of the family care environment. A high degree of each of these factors has been linked with a custodial care type of arrangement and a low degree of these dimensions has been associated' with a rehabilitative type of dwelling. Fifteen per cent of the homes scored above 80% on the level of restrictiveness and only 5% of the residences fell into the same category on the level of deviation from normative living (maximum score 100%), indicating that only a small minority of homes could be placed in a custodial care type category in relation to each of these dimensions.

Moreover, a sizeable proportion of homes, approximately 40% of the sample, scored under 50% on the level of deviation from normative living demonstrating that some opportunities are provided for residents to experience different facets of family living. The lack of association found between the level of restrictiveness and degree of deviation from normative living would seem to indicate that the pattern of care is less consistently structured than other dwellings dealing with a similar population. This inconsistency may be the result of a variety of "cross-pressures" on the sponsors related to the decentralized manner in which the program is administered. Sponsors who were previously employed in a custodial care setting scored significantly lower on the level of deviation from normative living scale than sponsors lacking such experience. Sixty-nine per cent of the "trained" sponsors as compared with 39% of the "untrained" sponsors scored below the median on the level of deviation from normative living (p < .01), indicating that the former group may be more involved in "rehabilitative" tasks than the latter group.

Younger sponsors appeared to maintain more restrictive homes than older caretakers. Sixty-seven per cent of the caretakers under 40 years of age scored above the median on the level of restrictiveness whereas only 29% of caretakers over 60 years of age scored above the median on the same dimension (p < .12), indicating that there may be more restrictions in homes managed by younger sponsors than in those managed by older ones. Comments obtained from the sponsors seem to suggest that the motivation of the individual for participating in the family care program may play an important role in shaping the structure of the setting. Such individuals as the older person who joins the program for companionship and the former employee of a custodial care facility who desires to take on a broader, more active role with psychiatric patients, may tend to establish settings consistent with these needs. What is noteworthy is the lack of relationship between size of the home and the social dimensions in the environment.

This negative finding brings into question a modification in family care programs which has been considered by some or already implemented by others, namely, limiting the number of ex-patients in a home. Based upon the findings different hypotheses were formulated to be tested in future research.