Validity of Perceived Stress Scale in Brazilian low-income college students

Abstract

OBJECTIVE: We tested the reliability and validity of the Perceived Stress Scale, an online questionnaire, among college students from low-income Brazilian regions. METHODS: We assessed 195 college students from a region with a Gini index of 0.56 for the validity study and a subsample of 117 students for the reliability study, where we evaluated the Perceived Stress Scale with the 14 original items. We also applied the shortened version of the Brief Symptom Inventory with 18 items (BSI-18). The psychometric properties analyzed, including temporal stability, internal consistency, and structural and convergent validity, were assessed using Spearman’s correlation coefficient, Cronbach’s alpha coefficient, unsupervised machine learning, and confirmatory factor analysis. RESULTS: The questionnaire showed acceptable reliability (temporal stability [rho ≥ 0.32] and internal consistency [alpha ≥ 0.79]. In construct validity, we identified two clusters, “helplessness” and “self-efficacy”, as structure solutions for our sample via unsupervised machine learning. An acceptable fit for the two-factor structure of the scale was indicated by multiple indices (chi-square/degrees of freedom [χ2/df] = 119/76; Tucker-Lewis Index [TLI] = 0.916; Comparative Fit Index [CFI] = 0.930; root mean square error of approximation [RMSEA] = 0.054; standardized root mean-squared residual [SRMR] = 0.078)) on confirmatory factor analysis. Moreover, convergent validity was supported by significant correlations of the BSI-18 Global Severity Index score with perception of helplessness (rho = 0.71) and self-efficacy (rho = -0.42). CONCLUSION: The Perceived Stress Scale, which is an online tool, is a reliable and valid self-report tool for college students.