Psychometric Properties and Measurement Invariance of the DERS-18 Scale in Rwanda

The study aims to provide validity and gender measurement invariance evidence for the DERS-18 (Victor & Klonsky, 2016), a short form of the Difficulties in Emotion Regulation Scale that maintains all the sub-domains included in the original 36-item version. The study addresses two relevant points raised by Vandenberg and Lance (2000). First, researchers frequently make inferences about differential group effects (e.g., treatment groups, gender comparisons) when testing statistical hypotheses, inferences that rely on measurement equivalence assumptions. Second, failing to meet these assumptions threatens the substantive, meaningful, and valid interpretation of statistical results. In the context of cross-cultural health and implementation sciences research in LMICs, this could have several practical implications for program evaluation, social programming, and funding. The primary purpose is to provide evidence of the statistical validity of a shorter version of the instrument in order to 1) contribute to strengthening the inferences about emotion regulation and mental health made in clinical and non-clinical research settings, 2) to reduce participants' burden when completing intervention/program assessments in Sub-Saharan Africa, and 3) to help reducing intervention costs and time in the context of the expansion of mental health-related programs, interventions, and global health research in LMICs. This main research objectives are: 1) Compare the DERS-18 internal consistency and latent structure against the 36-item DERS structure as defined by the original instrument developers (Gratz & Roemer 2004; Victor & Klonsky 2016); to 2) Evaluate DERS-18 gender invariance; and to 3) Assess convergent validity (correlations with anxiety, depression) and discriminant validity (group differences when comparing alcohol consumption) of the DERS-18.