Validating the Factor Structure and Measurement Invariance of the AUDIT-K Scale: Focusing on the Use of the RMSEAD Index for Nested Model Comparison

Abstract

This study was conducted to verify the validity and utility of the AUDIT-K by examining its factor structure and measurement invariances across gender and age groups, and longitudinal measurement invariance using data from the Korean Welfare Panel. In this process, RMSEAD, which has recently attracted attention, was used to verify any differences with the existing nested model comparison method. The data used in the analysis ranged from the 4th year to the 16th year. The main results of this study are as follows. First, AUDIT-K was found to be suitable for both two-factor and three-factor structures. Second, measurement invariance across gender and age groups was found to be maintained up to the weak invariance model. Third, longitudinal measurement invariance over 5 years at 1-year intervals was maintained up to the strict invariance model, and longitudinal measurement invariance over 13 years at 4-year intervals was found to be maintained up to the strong invariance model. Fourth, when comparing △RMSEA and RMSEAD for nested model comparison, it was confirmed that there were differences in results when comparing some models, suggesting that caution is needed in applying nested model comparison standards. Finally, the implications and limitations of this study were presented.