Using Dynamic Structural Equation Modeling to Examine Between- and Within-Persons Factor Structure of the DASS-21

Area of Science:

• Psychometrics
• Quantitative Psychology
• Clinical Psychology

Background:

• Traditional psychometric analysis often overlooks temporal dynamics.
• Integrating time series analysis with confirmatory factor analysis (CFA) allows for nuanced evaluation of measurement properties.
• Dynamic structural equation modeling (SEM) latent factor analysis (DSEM-CFA) offers a novel approach to assess instruments across individuals and over time.

Purpose of the Study:

• To apply DSEM-CFA to the Depression Anxiety Stress Scales-21 (DASS-21).
• To evaluate the reliability, invariance, and structural features of the DASS-21 subscales at both between- and within-person levels.
• To investigate the psychometric properties of the DASS-21 in a clinical sample undergoing regular therapy.

Main Methods:

• Utilized dynamic structural equation modeling (SEM) latent factor analysis (DSEM-CFA).
• Analyzed data from 333 individuals who completed the DASS-21 during regular therapy sessions.
• Examined autoregressive dependencies and psychometric properties at between- and within-persons levels.

Main Results:

• The DASS-21 demonstrated reliable measurement of depression, anxiety, and stress symptoms for between-person comparisons.
• Reliability was limited for assessing within-person fluctuations in symptoms over time.
• Current methods for modeling sensitivity to change may be insufficient.

Conclusions:

• DSEM-CFA provides critical insights into the reliability of measurement instruments both between and within persons.
• The DASS-21 is reliable for differentiating individuals but less so for tracking symptom changes within individuals over time.
• Further development of methods for assessing within-person change is warranted, especially for longitudinal studies.