Measurement Invariance and Construct Validity of the Satisfaction With Life Scale (SWLS) in Community Volunteers in Vietnam

Abstract

Worldwide, the Satisfaction With Life Scale (SWLS) has become the most widely used measure of life satisfaction. Recently, an authorized Vietnamese-language version has been introduced. Using a convenience sample comprising community volunteers from Ho Chi Minh City (N = 1073), confirmatory support was found for the cross-national constancy of the one-dimensional structure underlying the SWLS. Corrected item-total polyserial correlations and Omega coefficient were satisfactory. Using multi-group confirmatory factor analysis, configural, metric, and scalar invariance of the SWLS factorial structure were tested by gender, age, marital status, income, and educational level. Strong evidence of scalar invariance was found for gender and education, on which relevant subgroups did not differ in terms of latent means. Partial scalar invariance was found for marital status (item 4 and 5) and income (item 4). Being involved in an intimate relationship or having a higher income were associated with higher latent means. Scalar invariance in relation to age was very poor. Accordingly, caution must be exerted when comparing age groups. A high SWLS score was predictive of good self-rated health. Implications of the findings are briefly discussed.

Keywords: Satisfaction With Life Scale; Vietnamese population; confirmatory factor analysis; health status; income; marital status; measurement invariance; subjective well-being.

Figure 1

Path model for the Confirmatory Factor Analysis of the Satisfaction with Life Scale (SWLS), with factor loadings and residuals. Standardized coefficients are shown.

Figure 2

Path model for the association of the Satisfaction with Life Scale (SWLS) with the General Health item. Standardized coefficients are shown.