OBJECTIVETo develop the Chinese version of Children's Sleep Habits Questionnaire (CSHQ) by translating and retranslating the English version of CSHQ and to examine the reliability, content validity, construct validity of the Chinese version in a general population of school-age children.

METHODSDuring November and December 2005, a total of 20 457 school-age children from 55 primary schools of 9 cities were recruited through cluster-stratified selection, using geographic location, economic standard, and population density as criteria. A parent-administered questionnaire and the Chinese version of CSHQ were completed to quantify sociodemographic characteristics and to characterize sleep patterns and sleep disturbances, respectively.

RESULTSThe internal consistency of overall questionnaire and the eight subscales of the Chinese version of CSHQ was good (Cronbach's alpha coefficient was 0.73 and ranged from 0.42 to 0.69, respectively). The consistency between mother and father was ICCs = 0.89 for overall questionnaire and ranged from ICCs = 0.83 to ICCs = 0.92 for subscales. The test-retest reliability was ICCs = 0.85 for overall questionnaire and range from ICCs = 0.60 to ICCs = 0.88 for subscales. Factor analysis revealed a three-factor structure, which could represent bedtime behavior problems, sleep disturbance, sleep duration and daytime sleepiness, respectively, and the three factors could be used to explain 58.63% of the total variance. The factor loading was above 0.5 for the corresponding subscale.

CONCLUSIONSReliability, content validity and construct validity of the Chinese version of CSHQ in a general population of school-age children are good. It appears to be a suitable instrument for measuring sleep patterns and screening for sleep problems in school-aged children.