Psychometric Validation of the Sadaf Stress Scale-3 (SSS-3): A Multidimensional Measure for Pakistani Adults.

Abstract

Background: Stress is a multidimensional construct with significant health implications, yet most existing assessment tools are adapted from Western contexts and lack cultural validity for low- and middle-income countries. Earlier versions of the Sadaf Stress Scale (SSS) demonstrated promise for Pakistani populations but had methodological limitations, including reliance on internal consistency and absence of factor validation or empirical cutoffs. This study aimed to refine and validate the third version of the Sadaf Stress Scale (SSS-3) using modern psychometric standards, ensuring cultural specificity and methodological rigor in line with COSMIN and APA testing guidelines.

Methodology: A cross-sectional validation study was conducted among 820 adults (mean age = 29.8 ± 8.4 years; 50% females) recruited from universities, hospitals, workplaces, and community centers across Pakistan. The 50-item SSS-3 covered seven domains of stress (emotional, physical, traumatic, psychosocial, mental, nutritional, and chemical). Item refinement was achieved through expert review (S-CVI = 0.92) and cognitive interviews. Exploratory factor analysis (polychoric correlations, principal axis factoring) and confirmatory factor analysis (WLSMV estimation) were performed. Reliability was evaluated using Cronbach’s α, McDonald’s ω, split-half coefficients, and test–retest intraclass correlations (ICC). Validity was assessed through convergent, divergent, known-groups, and criterion analyses, followed by ROC-based cutoff derivation and normative data generation.

Results: EFA supported a seven-factor model explaining 64.1% variance; CFA confirmed strong fit (CFI = 0.958, TLI = 0.951, RMSEA = 0.046). Reliability was high across domains (α = 0.81–0.89; ω = 0.80–0.88; ICC = 0.76–0.84). Convergent validity was supported by correlations with PSS-10 (r = 0.69) and DASS-21 stress (r = 0.72); divergent validity with social desirability was weak (r = 0.10). Known-groups validity showed large group differences (d = 2.31, 95% CI: 2.02–2.60). ROC analysis yielded excellent discrimination (AUC = 0.87, 95% CI: 0.83–0.91) and empirically derived severity cutoffs.

Conclusion: The SSS-3 is a psychometrically robust, culturally tailored, and cross-culturally validated instrument for assessing stress among Pakistani adults. Its multidimensional structure, strong reliability and validity, and empirically defined cutoffs enhance its clinical, educational, and public-health utility.