OBJECTIVE The purpose of this study is to investigate the factor structure of the Center for Epidemiologic Studies Depression (CES-D) scale in community-residing stroke patients. METHODS The sample for this study was a convenience sample of 203 persons who had experienced strokes and were in a convalescent center for the disabled and a primary care unit in Korea. Data were collected through a questionnaire that included the CES-D scale. Cognitive function was assessed by the validated Korean version of the Mini-Mental State Examination (MMSE-K). The structure of the CES-D scale was analyzed by exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) using SPSS 17.0 and Amos 7.0. The reliability of the CES-D scale was assessed by calculating the internal consistency of Cronbach's α. RESULTS The 1-, 3-, and 4-factor models did not show the adequate fit indices. Two competing factor models were tested by CFA for the entire sample after identifying the factor structure using EFA. The 5-factor structure (loss of vitality, positive affect, psychomotor retardation, negative affect, and interpersonal problems) supported and explained 61.25% of the variance. CONCLUSIONS The results of this study suggest that the CES-D scale is a useful depressive symptom screening tool for community-residing stroke patients. Identifying the psychometric properties of the CES-D scale would help health professionals to understand the comprehensive assessment of community-residing stroke patients.