Measuring Perceived Social Support After Disaster: Development and Validation of the Perceived Social Support in Disasters Scale (PSSDS-28)

Abstract

Background/Objectives: Measuring perceived social support after disaster remains an unresolved psychometric problem: most instruments do not jointly capture the source of support (from whom it comes) and its function (how it operates), and none has been validated in a disaster-survivor sample with items framed around the retrospective appraisal of support received during recovery. We developed and validated the Perceived Social Support in Disasters Scale (PSSDS-28), a 28-item self-report measure integrating both dimensions through retrospectively framed items. Methods: In Phase 1, a PRISMA 2020-guided systematic review (489 records; 55 prioritized studies), a two-round Delphi with eight experts (content validity index = .996–1.000), cognitive interviews (n = 12), and a pilot study (N = 150) produced a 31-item form. In Phase 2, face-to-face data were collected from 607 adults affected by the 6 February 2023 Kahramanmaraş earthquakes (Türkiye). Factor-analytic, network-psychometric, item-response-theoretic, and common-method-variance analyses supported a five-factor, 28-item structure comprising family–relative, social environment, institutional, emotional, and instrumental–informational support. Results: The correlated five-factor model fit the independent validation sample well (CFI = .954, RMSEA = .076, SRMR = .072) and outperformed a one-factor alternative (ΔCFI = .222). Blocks showed strong internal consistency (ordinal α = .852–.949), composite reliability (.865–.950), and adequate test–retest stability (ICC = .542–.909). Scalar measurement invariance held across six groupings, with no meaningful differential item functioning. Convergent associations with general perceived support and loneliness, and negligible common-method variance, supported construct validity. Conclusions: The PSSDS-28 provides a defensible, multidimensional post-disaster assessment tool situated at the perceived–received interface that general-purpose scales are not designed to capture.