The Self-As-Carer Inventory was designed to permit individuals to express their perceived capacity to care for self. In the first phase of the study, a 44-item questionnaire was developed and tested through factor analysis primarily using a college population (Geden & Taylor, 1988). After the instrument was revised, a second study, reported here, was conducted with a more heterogeneous population. The age range of the respondents was broadened and the variability of health state and ethnicity was increased. Factor analysis with rotation was conducted on 589 completed inventories using the revised 40-item questionnaire. The 4-factor Promax solution accounted for 52% of the variance. Retention of factors with eigenvalues greater than 1 led to a 6-factor solution accounting for 58% of the variance. Significant positive correlations were found between total scores on the Self-As-Carer Inventory and ratings of health in general (r = .29, p = .0001), health at this moment (r = .25, p = .0001), and an estimate; of the amount of their own self-care provided (r = .36, p = .0001). No differences were found in total scores across health state, defined as respondents' reports of being sick-at-home (n = 130), hospitalized (n = 259), or well (n = 200).