INTRODUCTION: Studies of the 3-factor (family, friends and significant others) Multidimensional Scale of Perceived Social Support (MSPSS) have shown mixed results in non-Western and/or psychiatric populations due to factorial inconsistencies in its structure. Our study aimed to replicate and expand previous findings of MSPSS through confirmatory factor analysis (CFA) and measurement invariance analysis in a young Asian population of psychiatric and non-psychiatric subjects. MATERIALS AND METHODS: Data on 209 subjects were examined. The majority were Chinese (66.5%) followed by Malays (17.2%), Indians (14.4%) and other ethnicities (1.9%). Subjects in the non-psychiatric group (n = 100) did not report any psychiatric illnesses. Subjects in the psychiatric group (n = 109) were outpatients of a tertiary hospital in Singapore who had been diagnosed with depressive disorders. RESULTS: The 3-factor models of MSPSS showed better fit indices than the 2-factor models (friends/significant others and family, or family/significant others and friends) which indicated that the 3-factor structure of MSPSS was valid. Multigroup CFA demonstrated metric invariance, indicating MSPSS scores can be compared across groups. In the psychiatric group, descriptive and weighted univariate analyses revealed significantly lower levels of perceived social support in every domain of MSPSS. CONCLUSION: The 3-factor model of MSPSS can be used to compare psychiatric and non-psychiatric subjects locally. Since psychiatric patients reported lower MSPSS scores, future research could examine the causative factors that contribute to lower perceived social support in young adults seeking psychiatric intervention.
INTRODUCTION: Studies of the 3-factor (family, friends and significant others) Multidimensional Scale of Perceived Social Support (MSPSS) have shown mixed results in non-Western and/or psychiatric populations due to factorial inconsistencies in its structure. Our study aimed to replicate and expand previous findings of MSPSS through confirmatory factor analysis (CFA) and measurement invariance analysis in a young Asian population of psychiatric and non-psychiatric subjects. MATERIALS AND METHODS: Data on 209 subjects were examined. The majority were Chinese (66.5%) followed by Malays (17.2%), Indians (14.4%) and other ethnicities (1.9%). Subjects in the non-psychiatric group (n = 100) did not report any psychiatric illnesses. Subjects in the psychiatric group (n = 109) were outpatients of a tertiary hospital in Singapore who had been diagnosed with depressive disorders. RESULTS: The 3-factor models of MSPSS showed better fit indices than the 2-factor models (friends/significant others and family, or family/significant others and friends) which indicated that the 3-factor structure of MSPSS was valid. Multigroup CFA demonstrated metric invariance, indicating MSPSS scores can be compared across groups. In the psychiatric group, descriptive and weighted univariate analyses revealed significantly lower levels of perceived social support in every domain of MSPSS. CONCLUSION: The 3-factor model of MSPSS can be used to compare psychiatric and non-psychiatric subjects locally. Since psychiatricpatients reported lower MSPSS scores, future research could examine the causative factors that contribute to lower perceived social support in young adults seeking psychiatric intervention.