Development and psychometric validation of the 8-item Student Quality of Life Index (IQoL) using item response theory (IRT): a cross-sectional study among Brazilian higher education students.

objectivesTo develop and validate the 8-item Student Quality of Life Index (IQoL), a concise, multidimensional instrument for assessing quality of life (QoL) among higher education students in Brazil, encompassing psychological well-being, vitality and perceived functional capacity.

designCross-sectional psychometric validation study using a split-sample approach for exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), followed by bifactor item response theory (IRT) modelling and measurement invariance testing.

settingA large-scale survey conducted across 32 private higher education institutions in 14 Brazilian states, covering diverse academic fields. To ensure representativeness, the medical student subgroup was calibrated using post-stratification weights to align sex and age distributions with national medical education demographics.

participantsA total of 10 844 undergraduate students completed the survey. Psychometric model development used 10 793 respondents with complete data for the candidate item pool included in the EFA/CFA/IRT pipeline. Score distributions and subgroup comparisons used 10 838 respondents with complete information for sex, age group and course (3656 medical; 7182 other). The sample was predominantly female (69.0%) and white (47.3%) or mixed-race (41.2%), with an age range predominantly between 18 and 29 years. For medical-student comparisons, a stratified, calibrated analytic subset was created to match national sex and age margins, which led to a small reduction in the medical subgroup denominator in those analyses. PRIMARY OUTCOME MEASURES: Psychometric properties of the 8-item IQoL scale. Structural validity was assessed via CFA (weighted least squares mean and variance adjusted estimator) and internal consistency via Cronbach's alpha and Omega coefficients. Item-level performance was evaluated using Samejima's Graded Response Model. Scalar measurement invariance was tested across sex, age and course.

resultsThe parsimonious 8-item, three-factor model demonstrated superior fit indices (comparative fit index=0.996; Tucker-Lewis index=0.993; root mean square error of approximation=0.058 (90% CI 0.052 to 0.064); standardised root mean square residual=0.031). Internal consistency was high (α=0.88; [Formula: see text]). Bifactor IRT analysis supported a dominant general QoL factor, with item discrimination parameters ([Formula: see text]) ranging from 1.37 to 3.48. Scalar measurement invariance was established for sex, age and academic field, supporting valid group comparisons. Medical students reported slightly higher psychological well-being than non-medical peers, though effect sizes were trivial (Hedges' g=0.074). Within the medical subgroup, females scored higher in vitality and psychological well-being (g up to 0.225). Lower income and self-reported depression were significantly associated with lower global IQoL scores.

conclusionsThe Student Quality of Life Index (IQoL) is a psychometrically robust, invariant and efficient tool for large-scale monitoring of student well-being. The establishment of scalar invariance ensures that the observed differences across demographic and academic subgroups reflect true differences in the latent construct, reinforcing the instrument's utility for institutional assessment and mental health policy-making in higher education.