Measurement invariance and item functioning of the PHQ-15, DASS-21, and Paykel Suicide Scale across gender and cohorts in Peruvian university students

Abstract

This project hosts the supplementary material of an instrumental study on brief mental health screening measures used with university students. The study is a secondary analysis of two cohorts of an institutional screening program at a private university in Lima, Peru: 2020 (lockdown, n = 3,671) and 2024 (post-pandemic, n = 4,615). It examines the internal structure, reliability, item functioning, measurement invariance, and relations with other variables of the PHQ-15, the DASS-21, the Paykel Suicide Scale, and a brief ad hoc suicidality scale used in 2020. The analyses combine confirmatory factor analysis, Rasch and related item response models, ordinal multigroup invariance testing by gender and cohort, differential item functioning, and correlations among latent scores. The repository contains the supplementary tables of the article (Tables S1–S9). The database and the analysis scripts are not publicly shared. Therefore, they may be requested from the corresponding author.