Rasch analysis of the scores of the difficulties in emotion regulation scale (DERS) in a traffic context

Abstract

The functional approach of emotion regulation has been mainly used in the clinical context. However, other fields such as traffic psychology have also adopted this perspective, showing interest in how difficulties in emotion regulation might interfere in driving. Various driving styles have been associated with maladaptive emotion regulation. Difficulties in Emotion Regulation Scale (DERS) is the most commonly used instrument to assess ER difficulties and it is often used in the context of driving. Most studies have scored DERS by using procedures of the classical test theory approach instead of advanced psychometrics. In this study, a sample of 318 male drivers aged between 20 and 69 (M age = 41.6 years, SD = 11.0) completed the DERS. The total sample was divided in two halves: drivers with road traffic offences and matched controls. Data analysis was carried out using the Rasch Rating Scale Model (RSM). Results indicated that response categories performed inadequately. Collapsing the 5 original categories into 3 new categories was necessary. To meet the unidimensionality requirement, both the first DERS item and the awareness subscale were removed. Model-data fit was then good enough. Item Separation Reliability (ISR = 0.97) was excellent and Person Separation Reliability (PSR = 0.89) was quite good. Even though requirements of invariant measurement were met, there was low validity evidence for the driving context.