Validity and reliability of the Bermond-Vorst Alexithymia Questionnaire

The aim of this research is to establish the validity and reliability of the Bermond-Vorst Alexithymia Questionnaire (BVAQ). The BVAQ consists of five subscales, each comprising of eight items. The subscales are denoted Emotionalizing, Fantasizing, Identifying, Analyzing, and Verbalizing. The validity of the instruments was investigated by comparing results of psychometric analyses in three language groups. In addition BVAQ scores were correlated, for comparison, with the TAS-20 test scores and with measures of psychological problems. Two studies were carried out. In the first study the BVAQ was administered to a sample of Dutch students (n=375), a sample of French-speaking Belgian students (n=175), and a sample of English students (n=129). Cronbach's alpha was found to be about the same in the three samples (means of subscales about 0.79; total scales about 0.85). Principal components analyses of the 40 items revealed comparable five-factor solutions in the three samples. The mean total score of UK students differs about 0.29 SD from that of the Dutch and Belgian students. The intercorrelations of subscales were the same in the three samples. However, principal components analyses of the subscale scores of the English subjects revealed a different factor structure compared to that in the other samples. In the second study, the BVAQ and the Dutch version of the TAS-20 were administered to 430 Dutch students. Correlations between (sub)scales of the Dutch BVAQ and (sub)scales of the TAS-20 support the validity of the BVAQ. The validity of the BVAQ is further supported by correlations between BVAQ scores and measurements of psychological problems. (C) 2001 Elsevier Science Ltd. All rights reserved.