Structural Validity and Reliability of the Depression Anxiety Stress Scale in Peruvian University Students

Introduction: The Depression, Anxiety, Stress Scale (DASS-21) is a validated instrument in various populations; however, it needs to be continuously verified to ensure its correct use. Objective: Explore the structural validity and reliability of the Depression, Anxiety, Stress Scale (DASS-21), through different estimation methods in Peruvian university students. Method: An instrumental study was carried out. 406 health science students from a public university in Peru participated. Item variability, construct validity with exploratory factor analysis, the number of factors with Horn parallel analysis and confirmatory factor analysis were examined, including different estimation methods. Reliability and discriminative, correlational, predictive validity were analyzed. Results: A bifactorial structure of 21 items grouped into a general factor and three specific factors was corroborated. The model revealed optimal fit indices independently of the estimation method. The items discriminated satisfactorily, as well as the global scale according to age and religion. Correlation and predictive validity were adequate. The ordinal alpha was optimal for the global scale (0.969), and subscales depression (0.964), anxiety (0.890) and stress (0.980); Cronbach's Alpha, McDonald's Omega and Gutmann's coefficients supported these findings. Conclusions: The DASS-21 scale is a valid and reliable instrument to be used with Peruvian health science students. It is recommended to be used in research and clinical practice.