Psychometric properties of the perceived stress scale in Ethiopian university students

Abstract

Abstract Background Stress is a common psychological condition usually associated with many psycho-physical disorders. Stress and its risk factors are frequently seen in Ethiopians including university students. In such circumstances, a valid measure to screen for stress in Ethiopians is necessary. Therefore, we assessed the psychometric properties of the Perceived Stress Scale (PSS) in Ethiopian university students. Methods A cross-sectional study with a simple random sampling method was performed on students of Mizan-Tepi University, Mizan-Aman, Ethiopia. The study presents a psychometric investigation on a sample of 387 students (age = 21.8 ± 3.8 years, and body mass index = 20.8 ± 3.2 kg/m2) who completed PSS, Generalized anxiety disorder-7 scale (GAD-7), and a socio-demographics tool. McDonald’s Omega (internal consistency), factor validity for ordinal data and convergent validity (Spearman’s correlation) were assessed. Results No ceiling/floor effect was seen for the total or factor scores of the PSS-10 and PSS-4. Two factor model of the PSS-10 was favored by fit indices with Comparative Fit Index> 0.95, Weighted root mean square residual<.05 and root mean square error of approximation<.08. McDonald’s Omega was 0.78 and 0.68 for the PSS-10: Factor-1 and PSS-10: Factor-2, respectively. McDonald’s Omega was 0.70 and 0.54 for the PSS-4: Factor-1 and PSS-4: Factor-2, respectively. There were moderate-strong correlations (r = 0.62–0.83) between PSS factors and respective items loading on them. PSS scores were correlated with GAD-7 (r = .27–.40, p < .01). Conclusion The psychometric measures support the validity of the PSS-10 in Ethiopian university students.

Keywords

• PSS
• Anxiety
• Stress
• Africa
• Factor analysis,
• McDonald’s omega