Psychological Science Faculty Publications and Presentations

Construct validity and factor structure of the difficulties in Emotion Regulation Scale among adults with severe mental illness

Document Type


Publication Date




  • Assessed the psychometric properties of the DERS in a large SMI adult sample.

  • Construct validity was good to excellent with measures of psychiatric symptoms.

  • Five and six-factor intercorrelated models produced equivalent robust fit indices.

  • Ongoing research should utilize the original scale given extensive clinical utility.



The Difficulties in Emotion Regulation Scale (DERS: Gratz and Roemer, 2004) is a measure of emotion-regulation capacities with good construct validity, test–retest reliability and internal consistency. Factor analytic studies have produced mixed results, with the majority of studies supporting the original 6-factor model while several studies advance alternative 5-factor models, each of which raises questions about the psychometric validity of the Lack of Emotional Awareness factor. A limitation of prior psychometric studies on the DERS is the reliance on healthy subjects with minimal impairment in emotion regulation. The current study assesses the construct validity and latent factor structure of the DERS in a large sample of adult psychiatric inpatients with serious mental illness (SMI).


Inpatients with SMI (N = 592) completed the DERS, Acceptance and Action Questionnaire (AAQ-2), Patient Health Questionnaire (PHQ-SADS), and research diagnostic interviews (SCID I/II) at admission.


DERS total scores were correlated with AAQ-2 (r = .70), PHQ-Depression (r = .45), PHQ-Anxiety (r = .44) and moderately correlated with PHQ-Somatization (r = .28). Confirmatory factor analysis indicated that five and six-factor model produced equivalent fit indices. All factors demonstrated positive correlations with the exception of difficulty engaging in goal-directed behavior and lack of emotional awareness.


The DERS is a strong measure with excellent internal consistency and good construct validity. Caution is warranted in discarding the six-factor model given the equivalence with the five-factor model, particularly in light of the body of clinical research evidence utilizing the full scale.


Copyright © 2014 Elsevier Ltd. All rights reserved.

Publication Title

Journal of Psychiatric Research