Measurement invariance and cross-linguistic validation of the PSS-4 in university context: multidimensional analysis and associations with psychological and behavioral outcomes.

Background: Although prior research supports the use of the Spanish version of the PSS-4 using classical psychometric methods, further analysis of its dimensionality, reliability, and response patterns is warranted. Sociodemographic factors such as gender and health behaviors (e.g., sleep, diet, physical activity) may influence perceived stress. Objectives: This study aimed to: (1) evaluate the reliability and validity of the PSS-4 in Spanish, English, and, for the first time, Catalan; (2) test measurement invariance across language, university groups, and gender; and (3) examine associations between stress and psychological (depression, anxiety, wellbeing) and behavioral outcomes (sleep, internet use, physical activity). Methods: Participants included 1,810 students and 1,060 university staff, who completed surveys in Spanish, Catalan, or English. Demographic data included gender identity, marital status, education, and lifestyle behaviors. Measures included the PSS-4, WHO-5 Wellbeing Index, GAD Questionnaire, and PHQ. Dimensionality was examined using PCA, followed by invariance testing. The English group comprised a comparatively smaller sample. Results: The PSS-4 showed a unidimensional structure, high reliability, and strong correlations with psychological outcomes. Measurement invariance was supported at the configural and metric levels but not at the scalar level across languages, university communities, and gender. Conclusion: The PSS-4 demonstrates validity and reliability for assessing perceived stress in Spanish, English, and Catalan university settings, with this study providing the first validation of the Catalan version and a cross-language invariance test. However, the absence of scalar invariance limits the comparability of stress mean scores across languages.