Further Validation of the Chinese Short Warwick Edinburgh Mental Wellbeing Scale (SWEMWBS) in the Adult Population of Macau: An Application of Classic Test Theory and Item Response Theory

Lawrence T Lam ^1,2,3* Mary K Lam ⁴

• ¹ Faculty of Medicine, Macau University of Science and Technology, Macau, Macao, SAR China
• ² The University of Sydney, Darlington, New South Wales, Australia
• ³ Faculty of Health, University of Technology Sydney, Ultimo, New South Wales, Australia
• ⁴ RMIT University, Melbourne, Victoria, Australia

This study aims to validate the Chinese version of the Short Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS) by employing both Classical Test Theory (CTT) and Item Response Theory (IRT) approaches.Data were gathered through a population-based, cross-sectional health survey using an online self-reported questionnaire. The scale underwent Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). Measurement invariance by gender was assessed using standard procedures. The Grade Response Model (GRM) of the IRT analysis was applied to the data, estimating the discrimination and difficulty parameters at different thresholds. The results were analyzed both graphically and through parameter values.Factor analyses confirmed that a single-factor model of the scale fit the data well, with an overall Eigenvalue of 4.55, explaining 65.0% of the total variance. Model fit statistics were slightly better for males than for females. Measurement invariance examinations also yielded satisfactory Goodness-of-Fit statistics (CFI = 0.940, TFI = 0.910, RMSEA < 0.001) with minimal changes in item loadings and indicator threshold patterns across groups. The IRT results demonstrated high discrimination parameters, ranging from 2.17 to 3.67, and nearly evenly distributed difficulty parameters, ranging from -2.23 to 1.77. Graphical examinations indicated good performance of the scale across the latent trait continuum.The results indicated that, as a single-factor scale, the instrument exhibits good quality at both the scale and item levels. It has high discriminative power and an adequate response set for assessing a full range of the latent trait, namely mental well-being.