Psychometric Evaluation of the Multidimensional Scale of Perceived Social Support Among Early Adolescents in Darjeeling, India

Abstract

Background. The Multidimensional Scale of Perceived Social Support (MSPSS) is widely used, although it has not been validated among early adolescents in Darjeeling, India. The aims of the study were to validate the psychometric properties of the MSPSS, and to test for measurement invariance by gender. Methods. Survey data was collected from 274 early adolescents ages 10–14 living in Darjeeling, India. Confirmatory factor analysis (CFA) evaluated 1-, 2-, and 3-factor models. Reliability (Cronbach’s α, McDonald’s ω), convergent (peer problems), and concurrent validity (prosocial behavior) were assessed. Measurement invariance by gender was tested using multi-group CFA. Results. The three-factor model of the MSPSS (Family, Friends, Significant Other) fit these data well (X2[49] = 69.3, p = 0.030; CFI = 0.98; RMSEA = 0.039; SRMR = 0.036). Measures of reliability, concurrent, and convergent validity were good with MSPSS scores correlated positively with prosocial behavior and negatively with peer problems (|r| = 0.30–0.45, p ≤ 0.001). Configural invariance by gender was not supported, indicating differences in item-level loadings. Limitations. The MSPSS is a self-report measure, and social desirability bias is a potential limitation. Conclusion. The MSPSS demonstrates good reliability and validity among early adolescents in Darjeeling, India. Given non-invariance by gender, subscale comparisons across boys and girls should be interpreted with caution.