Psychometric properties of the Chinese (Mandarin) TEMPS-A: a population study of 985 non-clinical subjects in China.

Background: The TEMPS-A (Temperament Evaluation of Memphis, Pisa, Paris and San Diego) is a 110-item auto-questionnaire (self-rated) which consists of five temperament scales: depressive, cyclothymic, hyperthymic, irritable and anxious temperaments. It has been translated into over 25 languages and validated in at least 12, with broad cross-cultural cogency. This is a first attempt to validate the TEMPS-A in a very large Chinese population speaking Mandarin.

Methods: The Chinese TEMPS-A was adapted from the original English version following a rigorous process of forward translation and backward translation (after the approval of the English back translation by H.S.A. and K.K.A.), it was administered to 985 non-clinical Chinese subjects aged between 18-60 years (53.8% female) in four communities in Guangzhou City, China. A subset of 105 subjects was retested approximately six weeks later. Standard psychometric tests of reliability and validation were performed.

Results: The test-retest reliability for depressive (0.74), cyclothymic (0.71), hyperthymic (0.67), irritable (0.66) and anxious (0.83) were respectively as shown in the parentheses. For internal consistency, Chronbach alphas coefficients were 0.68, 0.85, 0.82, 0.83 and 0.87, respectively. Exploratory factor analysis revealed 2 super factors, Factor I loading on anxious, cyclothymic, irritable, and depressive temperaments; and Factor II loading on hyperthymic. Depressive, cyclothymic, irritable and anxious temperaments were correlated with each other. Males had significantly higher scores than females for the hyperthymic and irritable temperaments. The prevalence of the dominant depressive (2.9%), cyclothymic (5.6%), hyperthymic (1.3%), irritable (7.0%) and anxious (5.3%) temperaments were respectively as shown in the parentheses.

Limitations: Although it is likely that generalizability of our scale is good for the entire Mandarin-speaking ethnic composition of China today, future research is needed to establish this conclusively.

Conclusion: The Chinese TEMPS-A standardized on one of the largest non-clinical samples in any of the other national studies to date, has good internal consistency, coheres well with validated versions in other languages. The findings suggest that it is a psychometrically sound instrument to assess affective temperaments in clinical and biological studies in China.