Schizophrenic patients leaving a VA hospital for independent life in the community were followed up for a year after departure. They had been rated on ward and at activities, had been interviewed, and had been extensively tested. By 1 year 33 of 78 had returned, but there was little difference in the predeparture measures or in a 1-month home visit between these and those who stayed out. For those staying out, 35 of 111 predictors correlated significantly with at least 1 of 2 year-end adjustment criteria. Social assets predicted no better positively (and may have predicted worse), than psychopathology predicted negatively. Demographic variables, such as time since 1st mental hospitalization, predicted at least as well as tests; and factorially pure test scales predicted no better than empirical scales. (33 ref.) (PsycINFO Database Record (c) 2006 APA, all rights reserved)