LAUSR

Empathy corresponds to each one’s ability to understand other’s mental state, while self-differentiating from other. Most countries expect their health students and practitioners to be empathetic for people suffering from pathologies. The Interpersonal Reactivity Index (IRI) (Davis, 1980) and the Jefferson Scale of Empathy (JSE) (Hojat et al., 2002) are well-recognised scales designed for an estimation of general and clinical empathy, and usually used for between-group comparisons and analysis of time-related changes. In most comparative works, the conclusions were marred by methodological bias since measurement invariance (MI) which shows that item understanding is not influenced by group membership, remained to be demonstrated. This work aimed to analyse cisgender-related differences, after assessing MI of French IRI and JSE-student version (JSE-s) scales. It was carried out on a sample of health-trained newcomers who did not report psychological disorders. A multi-group confirmatory factor analysis was used for MI assessment. The dataset corresponded to 1216 questionnaires. The results confirmed the factor structure of both scales in cisgender groups considered separately. Non-invariant items across cisgender were identified in both scales. When invariant items were used for score calculation, it was observed that general and clinical empathy were significantly higher in cis women. To our best knowledge, this work presents the first MI analysis of French IRI and JSE-s scales. It proposes new findings and a robust methodology improving the reliability of empathy scales. The conclusions will help research and teaching teams for empathy analysis in broad populations, and for the training of health students.