Measurement Invariance of the Subjective Well-Being Scale in Low-Burden Settings: Psychometric Evaluation of the EBS-8 in Chile

To read the full-text Open Access article and download the PDFs, please visit our Issue portal at Paradigm (Walter de Gruyter): https://reference-global.com/article/10.56508/mhgcj.v9i1.376 Introduction: Subjective well-being (SWB) is a core indicator of psychological functioning across adulthood. In Latin America, brief and culturally appropriate measures are needed to support routine well-being assessment in community, primary healthcare, and other settings where lengthy protocols may reduce feasibility and participation. Purpose: This study aimed to examine the psychometric properties and measurement invariance of the Subjective Well-Being Scale (EBS-8) in Chilean adults. Methodology: An instrumental, cross-sectional study was conducted using non-probabilistic sampling. A total of 861 Chilean adults completed the EBS-8 and DASS-21; women represented 68.2% of participants, and the mean age was 26 years in both analytic samples. The analytical strategy included exploratory factor analysis, confirmatory factor analysis, reliability estimation using Cronbach’s alpha and McDonald’s omega, multi-group measurement invariance testing across gender at the configural, metric, scalar, and strict levels, and convergent validity analyses. Results: A unidimensional structure was supported, with high factor loadings and adequate fit (CFI = .989; TLI = .985; RMSEA = .081; SRMR = .011). Internal consistency was high (? = .981; ? = .983). Multigroup analyses supported configural and metric invariance between men and women. At the scalar and strict levels, the chi-square difference tests were statistically significant (p = .014 and p = .003, respectively), although the changes in CFI were small (|?CFI| = .002 in both models) and remained within conventional invariance criteria. EBS-8 scores correlated negatively with depression (r = –.471), anxiety (r = –.360), and stress (r = –.371). Conclusions: The EBS-8 exhibited a unidimensional structure, high internal consistency, and associations with psychological distress in the expected directions. The results supported configural and metric invariance between men and women. However, the significant chi-square difference tests at the scalar and strict levels, despite small changes in fit indices, indicate that conclusions regarding full measurement equivalence and comparisons of group means should be interpreted with caution. Its eight-item format may reduce respondent burden and facilitate application in community, primary healthcare, MHPSS, occupational, and population-monitoring contexts. These findings should be interpreted with consideration of the non-probabilistic, cross-sectional design.