| he purpose of this study was to explain the attitudes toward homeless persons and patients of health practitioners in training, including 116 medical students, 71 internal medicine residents, and 142 social work students. Males comprised 44 percent and females comprised 56 percent of the sample. The respondents ranged in age from 21 to 55 years with a mean age of 29.7 years.;Subjects completed a questionnaire developed specifically for this study. Variables measured included attitudes toward homeless persons and patients measured by a semantic differential scale, beliefs about disorders in the homeless population, quantity and quality of contact with homeless persons, locus of responsibility and locus of control beliefs, beliefs about dangerousness, beliefs about patients, willingness to help homeless persons, political attitudes, years of education, age, and gender.;Stepwise multiple regression was performed and a model consisting of four variables; quality of contact, locus of responsibility beliefs, beliefs about disorders, and locus of control beliefs, was found to predict 36 percent of the variance in attitudes toward homeless persons and patients. Additional analyses determined that attitudes toward homeless persons and patients differ significantly for respondents in medicine and social work (t(315) = ;Attitudes toward homeless persons and patients was significantly correlated with political attitudes (r(276) =.29, p... |