Abstract
Sleep is important for good physical and mental health. The COVID-19 pandemic lockdown created a unique context that impacted psychological and social drivers for sleeping well. The Pittsburgh Sleep Quality Index (PSQI) is a widely used measurement tool assessing subjective sleep quality. The traditional model of the PSQI (a one-factor model), whilst validated and used across different populations, has also been questioned with regards to data fit and representativeness of its global score in different social and work-related circumstances. Examination of the structure validity of the PSQI in the unique context of the pandemic has been scarce. This study determined the PSQI structure validity amongst employed women considered to experience increased stressors during the pandemic lockdown. The subjectively reported PSQI data from 498 female workers (mean age 44.6 years) collected during New Zealand's first national COVID-19 lockdown (April, 2020) was used. Confirmatory factor analyses compared the original one-factor model of the PSQI with the two- and three-factor models used by Jia et al. (2019) within this pandemic context. Results showed that the two-factor model provided a superior fit of the PSQI data compared to the original one-factor or a three-factor model. These findings suggest that a sub-score of the PSQI with two factors appears to be better at describing the sleep quality of healthy working women during the constrained situation of the pandemic lockdown compared to a single global sleep quality score. This indicates the importance of considering the validity of subjective sleep measures when used within unique social contexts and stressors.