Psychometric and structural properties of the four-item version of the Perceived Stress Scale: a replication study in 8727 school attendees

Objective. Though extensively used as a unidimensional measure, the four-item version of the Perceived Stress Scale (PSS-4) has unclear psychometric and structural properties. This study replicated past research by assessing the properties of both a one- and a two-factor model. Method. Relying on a nationally representative sample of 8,727 school attendees (MAGE = 14.8; SDAGE = 2.4; 48.9% female), the conducted analyses involved classical test theory and item response theory techniques. Results. Factor-analytic findings supported a two-factor model over a one-factor model, with the latter exhibiting poor-to-small scalability, monotonicity violations, and low reliability. Conclusion. The study findings suggest that the PSS-4 may not be purely unidimensional, which may call into question the use of its total score.