Abstract
INTRODUCTION: Life satisfaction is a core component of subjective wellbeing (SWB) that contributes significantly to child and adolescent development and outcomes. The Brief Multidimensional Students' Life Satisfaction Scale (BMSLSS) is one of the most popular measures of life satisfaction in childhood and adolescence. Most studies have validated this scale using classical psychometric methods. The purpose of this study was to thoroughly investigate the psychometric properties of the BMSLSS using Item Response Theory, in Chile. METHODS: We used Samejima's Graded Response Model (GRM) to investigate the item response functions, item and total precision, as well as gender and age differential item functioning (DIF) of the BMSLSS in a nationwide cross-sectional sample of n = 5,619 Chilean early adolescents (49.2% girls), aged 10, 11, and 12 years (46.13%, 44.99%, and 8.88% respectively). RESULTS: Conventional psychometric analyses, including item-rest correlations, internal consistency, and factor analysis, indicated good overall functioning of the BMSLSS. However, the IRT analysis revealed important nuances in the scale's performance. Specifically, results showed redundancy among lower response categories and insufficient discrimination at the upper end of the life satisfaction spectrum. The analysis also detected measurement non-invariance for some items across gender and age. DISCUSSION: These findings suggest that while the BMSLSS demonstrates adequate overall performance, the results advise reviewing the items' response options and exercising caution when interpreting high scores. Implications and future directions of research are discussed.