Validation of the Empathy Quotient (EQ) - Greek version.

Abstract

The original English language Empathy Quotient (EQ) is a self-reporting questionnaire that measures the construct of empathy in adults of normal intelligence. The EQ is sensitive to gender, and neurodevelopmental disorders. The EQ has been translated to many languages all over the world. The EQ - Greek version may be available through open access from www.autismresearchcentre.com. Aim of the present study was to validate the EQ- Greek version.The study took place in the 1st and 2nd Departments of Psychiatry of the National and Kapodistrian University of Athens (NKUA), Eginition and Attikon Hospitals respectively, and in the Korydallos Prison Psychiatric Clinic in Athens. Two groups completed the original 60 items version. One group consisted of general population and volunteer students from post graduate training programs (normal control group, N= 127) and the other group of patients recruited from the Adult Neurodevelopmental Disorders Unit of the 1st Department of Psychiatry of NKUA, the outpatients clinic of the 2nd Department of Psychiatry of NKUA and the Korydallos Prison Psychiatric Clinic (patient group, N=196). Three versions of the EQ were examined: the EQ-40, EQ-28 and EQ-15. All versions showed very good internal validity: Cronbach s a value was 0.902, 0.892 and 0.793 respectively. They all showed good test-retest variability: the Intraclass Correlation Coefficient was 0.928, 0.924 and 0.855 respectively. Concurrent validity examined by the correlation analysis with the Interpersonal Reactivity Index (IRI) showed non-significant correlations between the EQ and the IRI. Exploratory Factor Analysis (EFA) indicated a one-factor structure for the three versions. Confirmatory Factor Analysis (CFA) for the one-factor structure showed a good fit for all the three versions. CFA for the three-factor structures (Cognitive Empathy, Emotional Empathy, Social Skills) showed also a good fit for EQ-28 and the EQ-15. When the EQ-40 was used as a measure of empathy in a single dimension in adults, the EQ discriminated the normal control group from the patients group. The mean EQ score for the total sample was 35.84 with the lowest scoring being among Autism Spectrum Disorder (ASD) patients. As expected, females scored higher than males (p<0.001). To conclude, the Greek version of EQ showed good psychometric properties and could serve as a useful tool for clinicians to assess empathy in clinical populations and especially in subjects with ASD and other neurodevelopmental disorders.