The convergent, discriminant, and nomological validity of the Depression Anxiety Stress Scales-21 (DASS-21)

Abstract

BackgroundPrevious validity studies of Depression Anxiety Stress Scales-21 (DASS-21) through confirmatory factor analysis (CFA) were limited by yielding poor factor loading values and eventually skipping the critical process addressing convergent, discriminant, and nomological validity. Without the crucial process, it was risky that past studies insisted DASS-21 be ready for use. Therefore, the current study examined DASS-21 regarding the three kinds of validity to determine if DASS-21 can be further used in the future. MethodsThe current study formulated a CFA model (i.e., modified one-factor model) of DASS-21, which yielded factor loading values over β = 0.70 and estimated a latent variable of Psychological Distress by employing indicators of Depression, Anxiety, and Stress. The data collection was conducted at two universities in the U.S. and South Korea in November 2016 (N = 255). The study participants completed an online survey developed with DASS-21. With the collected data, the present study investigated the convergent, discriminant, and nomological validity of the modified one-factor model of DASS-21 through CFA. ResultsThe model fit of the modified one-factor model was complete as saturated. The convergent validity coefficient of Psychological Distress was 0.87, and average variance extracted of the variable was 0.68. The three correlations among the measured variables (i.e., indicators) were 0.50 < r < 0.75. LimitationsApplication of the current study's findings to youth or older adults can be limited as the subjects were young adults. ConclusionsThe current study made substantial evidence regarding the convergent, discriminant, and nomological validity of DASS-21 through CFA.