Validation of a Chinese version of the stress overload scale-short and its use as a screening tool for mental health status

Wenjie Duan,Wenlong Mu
DOI: https://doi.org/10.1007/s11136-017-1721-3
Abstract:Purpose: Although stress emerges when environmental demands exceed personal resources, existing measurement methods for stress focus only on one aspect. The newly-developed Short Stress Overload Scale (SOS-S) assesses the extent of stress by assessing both event load (i.e., environmental demands) and personal vulnerability (i.e., personal resources). The present study was designed to evaluate the psychometric properties of the Chinese version of Stress Overload Scale-Short (SOS-SC), and further examine its roles in screening mental health status. Methods: A total of 1364 participants were recruited from communities and colleges for scale validation. Results: Reliabilities were good throughout the subsamples (ω > 0.80). Confirmatory factor analysis indicated the acceptable goodness-of-fit for the two-factor correlated model (Sample 1: 560 community residents). Multi-group confirmatory factor analysis confirmed measurement invariance across community residents (Sample 1) and college students (Sample 2 and Sample 3). Criterion validity and convergent validity were established (Sample 2: 554 college students). Latent moderated structural equations demonstrated that the relationship between SOS-SC and depression is moderated by social support (Sample 2), further validating the SOS-SC. In addition, the SOS-SC effectively screened individuals in a population at different levels of mental health status (i.e., "at risk" vs. "at low risk" for depression symptoms and/or wellbeing). Conclusion: The SOS-SC exhibits acceptable psychometric properties in the Chinese context. That said, the two aspects of stress can be differentiated by the Chinese context, therefore, the SOS-SC can be used to measure stress and screen mental health status among the Chinese population, and monitor and evaluate health-promoting interventions.
What problem does this paper attempt to address?