Towards a new structure of the interpersonal reactivity index. reliability and validation of the Portuguese version: A comparative analysis

Abstract

ObjectiveEmpathy has received a lot of attention with the creation of an Interpersonal Reactivity Index (IRI). IRI is assessed using a 28-item questionnaire with four 7-item scales:– perspective-taking (PT) scale;– fantasy (FS) scale;– empathic concern (EC) scale and iv) personal distress (PD) scale.Such questionnaires have been translated in many languages (see Table 1). This work aims to compare the original model with existing modified, hierarchical, shortened models in the literature and to also translate and validate the IRI for Portuguese.MethodsA convenience sampling was used (n = 130) to conduct confirmatory factor analysis (CFA) using AMOS software. The original four-factor model is contrasted with modified, hierarchical and shortened models proposed in the literature, reporting internal consistency statistics and their fit indices.ResultsThe same structure of factors was found in the sample with reasonably good fit indices χ2/df = 1.57, CFI = 0.77, AGFI = 0.72 and RMSEA = 0.067. Internal reliability for each scale of the IRI was not excellent (< 0.90), but it is in line with the literature: PT with a Cronbach's alpha of 0.74, FS with 0.79, EC with 0.74 and PD with 0.65. The comparison with other modified versions of the IRI latent factor structure revealed that two models with better fit than the original version, and the potential for a shortened Portuguese version of the IRI.ConclusionIRI is a valid instrument to measure empathy in the Portuguese Population and is in line with previous findings. Some modifications to the original latent structure provide a better data fit than the original one.Disclosure of interestThe author has not supplied his/her declaration of competing interest.