Psychometric properties of the Epworth sleepiness scale in Ethiopian university students

Abstract

Background: Daytime sleepiness is highly prevalent across the globe. The Epworth sleepiness scale (ESS) is the most widely used tool for screening daytime sleepiness. The psychometric properties of the ESS have not been comprehensively examined in African populations. Material and methods: A cross-sectional design with simple random sampling was used in the present study. The study recruited 600 students from Mizan-Tepi University, Ethiopia, of which 329 (age = 18-28 years and body mass index = 21.19 ± 3.17 kg/m 2 ) completed the study. ESS, a semi-structured socio-demographics questionnaire and a clinical interview to diagnose insomnia according to the International Classification of Sleep Disorders were employed. Results: All except one item of the ESS showed a floor effect, while only one item score showed ceiling effect. However, no ceiling/floor effect was observed in the ESS total score. The Cronbach's alpha (0.75) and composite reliability (0.75), indicated good internal consistency, while a moderate item-total score correlation (r = 0.55-0.67) implied favorable internal homogeneity. The known-group validity was established by significantly higher scores for all the ESS item scores and the ESS total scores among those with symptoms of insomnia than among non-symptomatic students. Fit indices along with the consideration of inter-factor correlation coefficient, measures of item retention favored the unidimensional structure of the ESS. Conclusion: The ESS has excellent psychometric validity for screening daytime sleepiness in Ethiopian university students.