Construct validity of brief difficulties in emotion regulation scale and its revised version: evidence for a general factor

The Difficulties in Emotion Regulation Scale (DERS) was developed to assess the multidimensional structure of emotion dysregulation. This study investigated the validity of a brief version for the DERS (DERS-18) and its revised version (DERS-15). We proposed a bifactor model for representing the unidimensional construct underlying the multi-facets of emotion dysregulation and examined the fit of the bifactor structure for DERS-18 and DERS-15 by estimating and comparing a series of measurement models suggested by previous studies, using a non-clinical sample of 907 Chinese college students. The bifactor model fits the data best compared to other competing models for DERS-18. Findings indicated that the majority variance of the clustered manifestations of emotion regulation deficits could be explained by one general emotion dysregulation factor, supporting the use of the total score of the DERS-18. The specific factors Awareness, Clarity, Goals, and Non-acceptance in DERS-18 demonstrated their specificity after controlling the effect of the general factor. The DERS-18 demonstrated good composite reliability, test-retest reliability, discriminant validity, and good criterion validity when connected with regulatory emotional self-efficacy and wellbeing indicators. The DERS-15 showed equally good model fit and reliability, and a small diminution in discriminant and criterion validity. Implications of the findings are discussed.