Cross Validation of the Korean Translation of Composite Scale(KtCS) to Measure Morningness-Eveningness

Kook, Seung Hee; Yoon, Jin Sang; Lee, Hyung Young

Abstract

OBJECTIVES: It is necessary to develop a questionnaire to classify the circadian rhythm for studying circadian rhythms and predicting shift work adaptability. This study attempted to confirm the general applicability of "Korean translation of Composite Scale(KtCS)" by Jin-Sang Yoon et al(1997), a Korean version of Composite Scale(CS) by Smith et al(1989).
METHODS
KtCS and Life Habit Questionnaire(LHQ) were administered to 155 nurses(85 rotating shift nurses, 70 nurses on a fixed day schedule) and 247 female university students. We tested the distribution of KtCS scores and then subjects were categorized on the basis of their KtCS scores into evening(E), intermediate(I), and morning(M) types. Cut-off points for these categories were set at the 10th and 90th percentiles. We obtained the results of the internal consistency and the structure of factors of KtCS. We also compared the differences of life habits according to the circadian rhythm types.
RESULTS
In both groups, nurses and female university students, the distributions of KtCS scores were negatively skewed but not significantly and the KtCS mean score was not significantly different. In nurses, Cronbach's alpha was 0.836, and three extracted factors were 'preferred times of performance and going to bed', 'rising time' and 'morning alertness'. In female university students, Cronbach's alpha was 0.787, and the same three factors as nurses were confirmed. However, only item 11 was loaded to factor 1 for the nurse group, whereas it was loaded to factor 2 for the student group. Life habits significantly differed among the three circadian rhythm types. The time of perceived peak alertness(mental and physical) was the latest in the E-type. The quality of academic performance during the morning was the highest in the M-type, and during the evening it was the highest in the E-type, but during the afternoon there was no significant differences among the three types. The bedtime and the rising time were latest in E-type, but the sleep lengths were not significantly different among the three types in both groups.
CONCLUSIONS
The psychometric properties of KtCS in nurses were very similar to those of female university students and the previous findings of Jin-Sang Yoon et al(1997). This result supports that KtCS can be generally applied.