Validity, reliability, measurement invariance, and equipercentile linking of the depression anxiety stress scale-21 in the youth population in Singapore

Abstract

Objective: The current study aims to examine the structural and convergent validity, reliability, and measurement invariance of the Depression Anxiety Stress Scales Short Form (DASS-21) across age and gender among the youths in Singapore. Additionally, it aims to provide a simple and reliable method for converting the DASS-21 Depression and DASS-21 Anxiety to the Patient Health Questionnaire 8-item (PHQ-8) and Generalized Anxiety Disorder 7-item (GAD-7) scores using an equipercentile linking method.

Methods: A total of 2,600 respondents were recruited from a National Youth Mental Health Study.

Results: The confirmatory factor analysis (CFA) confirmed that the original three-factor model fits our data. Cronbach's alpha coefficients for the depression, anxiety, and stress subscales were 0.91, 0.87, and 0.89, respectively. Multiple CFA across age and gender showed that the configural, metric, and scalar measurement invariance models strongly support the three-factor model. The intraclass correlation coefficient (ICC) between the raw and converted PHQ-8 and GAD-7 scores support that the DASS-21 subscale scores are practically exchangeable with the PHQ-8 and GAD-7.

Conclusion: These findings suggest that the DASS-21 is a valid tool for measuring depression, anxiety, and stress among the youths in Singapore.

Keywords: DASS-21; depression anxiety stress scale; equipercentile linking; reliability; validity.