Validity and reliability of the Difficulties in Emotion Regulation Scale Short Form in Indonesian non-clinical population

Abstract

IntroductionEmotion dysregulation is a transdiagnostic marker for vulnerability and has high comorbidity rates across various range of disorders among adolescents and young adults, highlighting the crucial need for precise assessment tools to recognize its significant impact on well-being. The Difficulties in Emotion Regulation Scale Short Form (DERS-SF) is a comprehensive instrument designed to measure the ability to regulate emotion. This study aimed to investigate the psychometric properties of DERS-SF among the non-clinical population, particularly high school and university students in Indonesia.MethodsA total of 738 senior high school and university students completed the Indonesian version of DERS-SF and standard questionnaires to assess its validity, consisting of the Depression Anxiety Stress Scale (DASS), the Beck Depression Inventory-II (BDI-II) for young adults and the Children Depression Inventory (CDI) for adolescents. Three models were examined in factorial validity tests using confirmatory factor analysis.ResultsThe results showed that DERS-SF had an overall good internal consistency with Cronbach’s alpha coefficient of.89 for the 18-item version,.90 for the 17-item version, and.91 for the 15-item version. Test-retest reliability was moderate with a value of.67. In addition, it had good satisfactory content as shown by item content validity index (I-CVI) = .96 and scale content validity index (S-CVI) = .83, as well as convergent validity. All subscales scores showed a positive and strong correlation with DASS, BDI-II, and CDI except awareness. Based on confirmatory factor analysis (CFA), the correlated 6-factor model excluding item number 6, and the 5-factor model excluding awareness were suitable to use in non-clinical populations.ConclusionThis study established the removal of the awareness subscale in the DERS-SF Indonesian version, resulting in better reliability and validity than the original version with complete subscales.