Investigating the Factor Structure of the Japanese Version of the Interpersonal Reactivity Index

The Interpersonal Reactivity Index (IRI) is widely used to measure empathy; however, the factor structure and scoring of the IRI have been inconsistent. Previous studies showed a bifactor model with empathic concern (EC) and perspective taking (PT) as a better model for the Chinese version of the IRI. This study investigated whether this model was better than other candidate models for the Japanese version of the IRI. In Study 1, we compared the model fit parameters of confirmatory factor analysis using datasets from previous studies. The results showed that the bifactor model with EC and PT was the best among 42 models. Study 2 investigated which concepts were extracted in a bifactor model that was derived from web‐based survey data. Results showed that the general factor in the bifactor model might measure the compound characteristics of both EC and PT. Additionally, the extracted EC factor might measure avoidance of harm to others, and the extracted PT factor might reference inferring others' minds. This study suggests that the bifactor model with EC and PT has the best factor structure and that the extracted factors reflect different aspects of empathy.