A Pilot Study Examining the Factor Structure of Pittsburgh Sleep Quality Index in Indian Context

The Pittsburgh Sleep Quality Index (PSQI) has gained widespread acceptance as a useful tool to measure subjective sleep quality. The Cantonese version of PSQI has shown the two factor model of PSQI fits all the data nicely. There is no such factor structure created for any Indian context to the best of our knowledge. In this work, we seek to estimate such a structure in the Indian context. We have used Confirmatory Factor Analysis (CFA) on the factors suggested in literature and endorsed in the Cantonese study. We found the two factor model fitting the data nicely for many cases (age groups) but it fails for other cases (or age groups). We attribute this to the somewhat small sample data size. Seeing this deficiency, we used the Exploratory Factor Analysis (EFA) technique to identify new factor structures. Using this new structure we found the data fitting better as is evident from a better Goodness of Fit index (CFI) in many cases. For the remaining cases the CFI is poor requiring these cases to be explored further.