© 2017 Informa UK Limited, trading as Taylor & Francis Group Objective: Difficulties in modeling the Multidimensional Scale of Perceived Social Support (MSPSS) have occurred, and these were corrected in the revised version of the Thai MSPSS. However, the revised version has not been tested in elderly populations. The present study aimed at confirming the factor structure of the revised version of the MSPSS among the elderly with depressive disorders, in populations with or without depressive disorders. Methods: Eight hundred and three elderly patients were recruited from four tertiary hospitals; 190 (23.7%) had depressive disorders. All completed the revised Thai MSPSS consisting of 12 items, using a 7-point Likert scale. Confirmatory factor analysis (CFA) of the MSPSS was conducted in both groups. Results: The mean age was 69.24 years (SD 6.88), and 70% of the sample was female. There were no significant difference in demographic data between two groups. The revised version of the Thai MSPSS provided excellent internal consistency. The three-factor model was clearly superior to other alternative models in both depressed and non-depressed groups. CFA for the whole group revealed an acceptable model fit: χ 2 = 147.44, df = 45, p < 0.001; Tucker-Lewis Index 0.975; Comparative Fit Index 0.982; Good Fit Index 0.966; and root-mean-square error of approximation 0.056. The fit statistics in the depressed group was better than in the non-depressed group across all models. Conclusions: Due to its robust factor structure, these data support the use of the revised MSPSS as a brief instrument for assessing perceived social support in the elderly with or without depressive disorders.