Development and evaluation of a Chinese translation of the State-Trait Anxiety Inventory and the Beck Depression Inventory

The State-Trait Anxiety Inventory (STAI-Y) and the Beck Depression Inventory (BDI) were translated into Chinese for use by Hong Kong Chinese. The translation process involved forward and backward translations and evaluations of linguistic and psychological content by bilingual professionals in clinical psychology, psychiatry and linguistics. The evaluation of cross-cultural equivalence of the translated versions was achieved through the administration of the original and the translated Chinese forms of the two instruments to 80 Chinese-English bilingual and bi-cultural Hong Kong Chinese in the Greater Los Angeles area. For the two anxiety scales of STAI-Y and BDI, high item correlations between the English and Chinese forms, similar and high item-total correlations, and similar high alpha coefficients of each scale provided evidence for their translation equivalence. The evaluated Chinese forms of these two instruments were then administered to 295 Chinese monolingual college students in Hong Kong to examine factor structures and to obtain a tentative non-clinical population norms on which means, standard deviations, and item-total correlations were obtained. T-score tables were presented. Exploratory and multiple group factor analyses were used to identify and compare to the factor structures of the STAI and BDI between the Hong Kong samples and the previous factor analyses. The procedure of the factoral analyses was based on Gorsuch's theory and statistical program on factor analysis. From the exploratory analysis for the STAI, (with the State and Trait scales factored separately), results demonstrated a two-factor solution for both the State and Trait scales. The factors identified were related to the presence of positive feelings/cognitions or the presence of anxiety symptoms (State A+, State A{dollar}-{dollar}, Trait A+, and Trait A{dollar}-{dollar}). The multiple group analysis found that the two-factor structure was invariant across sex and across age to an American college sample and a high school sample from the STAI-s manual. For the BDI, only one factor was identified and invariance was not found across five American and cross-cultural studies.