Cross-cultural adaptation and psychometric evaluation of the isiZulu 20-item Toronto Alexithymia Scale : preliminary findings.

Abstract

According to prevalence studies, 1 in 3 South Africans will present with a diagnosable mental disorder throughout the course of their lifetime, whilst over a given period of 12 months, 1 in 6 South Africans will likely present with a mental disorder that is clinically diagnosable. Given the alarming rates of mental illness in South Africa much attention within the literature has focussed on further understanding both its determinants and associated risk factors. Due to its significant associations with several psychiatric and medical conditions, the subject of alexithymia has advanced these aims in the global literature. However, little research on the subject has been undertaken in South Africa, most likely due to the lack of a validated psychometric assessment for detecting and further investigating the issue of alexithymia among patients. Additionally, research also indicates that the measurement of alexithymia is impacted by sociocultural factors that shape it in part. Consequently, this research aimed to address these issues by undertaking a cross-cultural adaptation of the 20-item Toronto Alexithymia Scale (TAS-20) for an isiZulu-speaking sample of university students and evaluating its psychometric properties. This was achieved through three objectives: (1) cultural adaptation and translation of the TAS-20 into isiZulu; (2) evaluating the psychometric properties of the isiZulu translation of the TAS-20; and (3) exploring variation in scores on the isiZulu TAS-20 by gender. The results of the confirmatory factor analyses suggested best global fit for the standard oblique three-factor model with an additional specified covariance between two indicators. Evaluation of local fit supported meaningful parameter estimates, as well as the convergent and discriminant validity for the DIF and DDF latent factors of the model, but failed to support the validity of the EOT latent factor. Reliability analyses similarly demonstrated acceptable reliability for the total isiZulu TAS-20 scale and the DIF and DDF subscales, but failed to do so for the EOT subscale. Lastly, there was tentative indication of significantly elevated levels of alexithymia among female Zulu participants as compared to male Zulu participants. In conclusion, the psychometric evaluation confirmed the factor structure, but failed to fully support the underlying theoretical relationships in the Zulu culture. In convergence with other studies, these findings suggest that sociocultural factors significantly impact upon the construct and assessment of alexithymia. The study suggests the need for a theoretical reconceptualization of alexithymia, factoring in the role of sociocultural factors.