Infant temperament at 6-9 months of age: validity of the French version ofthe Infant Characteristics Questionnaire and factors associated with the measurement

Background: The aim of the study was to assess the validity of the French v ersion of the Infant Characteristics Questionnaire (ICQ), which was develop ed by Bates in 1979 to investigate the parental perceptions of infant tempe rament at 6 months of age. Methods: The French version was established after translation/back-translat ion, and tested on a sample of 794 mothers, who gave birth in 1995 in four French Departments and who returned the questionnaire which was mailed 6 mo nths after the birth. Results: The non-response rate of each item was very low (<2%). Factor anal ysis resulted in the same four factors as in the American data: fussy/diffi cult, unadaptable, unpredictable, dull. However, some of the items did not have similar factor loadings. Thus new factor scores were defined for the F rench version of the ICQ. Internal consistency, as measured by the Cronbach coefficient, was satisfactory (> 0.7) for the factors fussy/difficult and unadaptable. It was lower, but acceptable (> 0.6) for the two other factors . The infant's temperament was not related to birth parameters. In contrast , some factor scores were significantly associated with parity, mother's ed ucation, as well as her health and the infant's health after leaving the ma ternity ward. Conclusions: The French version of the ICQ is well accepted and has a good validity. This scale provides a useful instrument for research, especially for epidemiological studies on infant health and development, as well as th eir determinants.