Applying generalizability theory to the Perceived Stress Scale to evaluate stable and dynamic aspects of educators’ stress

The 10-item Perceived Stress Scale (PSS-10) is the most widely used measure of perceived stress, yet its ability to distinguish between enduring and dynamic aspects of stress and temporal reliability have not been yet investigated using appropriate methodology. Generalizability theory is suitable for this purpose and was applied to investigate psychometric properties of the PSS-10. A longitudinal measurement design was used with data of 122 educators randomized into Intervention (mindfulness) and Control group and assessed on three occasions, baseline and after 8- and 6-week intervals. The PSS-10 demonstrated strong temporal reliability with assessment scores generalizable across persons and occasions (G = .86). Self-confidence and irritation were identified as the most dynamic/amendable aspects of perceived stress, which has implications for interventions aiming at efficient stress reduction but did not affect temporal reliability of the scale. This study demonstrated that the PSS-10 is a trait measure suitable to evaluate long-lasting effects of interventions aiming to reduce perceived stress. Researchers using the PSS-10 to assess effects of interventions should be aware that the scale might not be sensitive enough to reflect short-term intervention effects and is less suitable for this purpose. The current application of G Theory to PSS-10 highlights the need to develop a measure to assess perceived stress as a state and informs interventions aimed at effective stress reduction.