Psychometric properties of the Depression, Anxiety, and Stress Scale (DASS-21) among different Chinese populations: A cross-sectional and longitudinal analysis

• Published in: Acta psychologica Vol. 240; p. 104042
• Main Authors: Chen, I-Hua, Chen, Chao-Ying, Liao, Xiao-ling, Chen, Xiu-mei, Zheng, Xia, Tsai, Yu-Cheng, Lin, Chung-Ying, Griffiths, Mark D., Pakpour, Amir H.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2023; Elsevier
• Subjects: Anxiety and Stress Scale; Confirmatory factor analysis; DASS-21; Depression; Depression, Anxiety and Stress Scale; Factor structure; Longitudinal study; Measurement invariance; Confirmatory factor analysis; Factor structure; Depression, Anxiety and Stress Scale; Measurement invariance; DASS-21; Longitudinal study
• DOI: 10.1016/j.actpsy.2023.104042
• ISSN: 0001-6918; 1873-6297

Given that there is limited evidence concerning the psychometric properties of DASS-21 when applied to primary school students, the present study undertook a comprehensive exploration of the psychometric evidence supporting the use of the DASS-21 within this demographic. The research comprised three studies. In Study 1, the basic psychometric properties of internal consistency and construct validity were examined. A total of 3138 primary school students from three provinces in mainland China participated. The internal reliability of the overall scale was 0.93, and for all the subscales, it was higher than 0.80. Construct validity was partially supported. Both exploratory and confirmatory factor analyses upheld the factorial validity of the original three-factor structure. While convergent validity was established, the results showed unsatisfactory discriminant validity. The bifactor model showed that DASS-21 raw scores predominantly indicated the general factor, evidenced by the high explained common variance and omega-hierarchical values. However, the contributions from the three specific factors were minimal, with their omega hierarchical values all below 0.15. In Study 2, a longitudinal design was adopted, tracking 1366 primary school students from Southwest China over a three-month interval. The results further confirmed that the DASS-21 exhibited scalar time-invariance. The latent mean analysis showed that there were no statistically significant differences in the latent means of depression, anxiety, and stress between Time 1 and Time 2. In Study 3, which included 364 college students and 483 enterprise workers, the results demonstrated that the DASS-21 had measurement invariance across different populations. The latent mean analysis further confirmed that, in terms of the latent mean of all three subscales, both college students and enterprise workers had significantly higher scores than primary school students. Overall, the findings indicated that the DASS-21 is a suitable tool for screening schoolchildren for general psychological distress, but it is not suitable for discerning distinct negative mood state disorders.