A confirmatory factor analysis of the Toronto Alexithymia Scale (TAS-20) in an alcohol-dependent sample
Date of this Version
Objective: To explore the factorial validity of the Toronto Alexithymia Scale (TAS-20) in an alcohol dependent population. Methods: Confirmatory factor analyses were conducted using structural equation modelling to evaluate the stability of the factor structure of the TAS-20. Based on previously reported factor structures, 1-factor, 2-factor and 3-factor models were examined, using maximum likelihood estimation. Results: Four of the five fit indices considered indicated a superior fit for the 3-factor model compared to the 2-factor and 1-factor models. Conclusion: This study demonstrates that the original 3-factor model of the TAS-20 was a more plausible fit to the data compared to other competing models in a sample with alcohol dependence. Limitations with the EOT factor indicate a need for refinement and a possible revision of the TAS-20 in populations with alcohol dependence.
This document has been peer reviewed.