Dimensionality of the Pittsburgh Sleep Quality Index in the young collegiate adults

PURPOSE: To explore and validate the factor structure of the Pittsburgh Sleep Quality Index (PSQI) in young collegiate adults. METHODS: Six hundred university students were initially contacted and invited to participate in a survey of their sleep experience and history. Of this preliminary sample 418 of the students (age = 20.92 ± 1.81 years, BMI = 23.30 ± 2.57 kg/m(2)) fulfilled the screening criteria and ultimately completed the Pittsburgh Sleep Quality Index (PSQI), a self-report survey of respondents’ sleep habits and sleep quality. The students were enrolled in various undergraduate and postgraduate programs at Jamia Millia Islamia, New Delhi, India. Exploratory factor analysis (EFA) investigated the latent factor structure of the scale. Confirmatory factor analysis evaluated both of the models found by EFA. RESULTS: The Kaiser’s criteria, the Scree test, and the cumulative variance rule revealed that a 2-factor model accounted for most of the variability in the data. However, a follow up Parallel Analysis found a 1-factor model. The high correlation coefficient (r = 0.91) between the two factors of the 2-factor model and almost similar values of the fit indices supports the inference that the PSQI is a unidimensional scale. CONCLUSIONS: The findings validate the 1-factor model of the PSQI in young collegiate adults.