Factor structure of a Cantonese-version Pittsburgh Sleep Quality Index

Abstract

The objectives of this study were to translate, validate and test the factor structure of a Cantonese version of the Pittsburgh Sleep Quality Index (Cantonese-PSQI) in support of its use among Cantonese-speaking Chinese in Hong Kong, southern China and Chinese immigrant communities in the West.

The study involved Cantonese-speaking Chinese aged 45 or over in Hong Kong. A cross-sectional survey of 97 participants was firstly conducted to validate the Cantonese-PSQI. A second survey of 794 participants identified the best-fitted factor structure of the scale. Confirmatory factor analysis then functioned to develop and test a single-factor model, a two-factor model, and a three-factor model. The one-factor model turned out to be unacceptable due to its poor fit (L² = 303.408, d.f. = 14, SRMR = 0.085; RMSEA = 0.161; CFI = 0.805). The three-factor model, despite its very good fit (L² = 36.450, d.f. = 11, SRMR = 0.022; RMSEA = 0.054; CFI = 0.983), was unacceptable because of the extraordinarily high correlation (r = 0.928) between two of its three factors, namely the sleep quality factor and the daily disturbances factor. The two-factor model showed a good fit with the data of all the participants (L² = 46.339, d.f. = 13, SRMR = 0.028; RMSEA = 0.057; CFI = 0.978), and with those of the middle-aged, the young-old and the old-old groups. The two-factor model of the Cantonese-PSQI characterized by sleep efficiency and sleep quality is therefore better than the original one-factor model and the three-factor model for measuring the sleep quality of middle-aged and older Cantonese-speaking Chinese.