Assessment of community resilience frameworks: a validation through community outcomes after disaster events

Published

December 10, 2025

Author(s)

Donghwan Gu, Maria Dillard, Michael Gerst

Abstract

Resilience indices have become increasingly popular tools to assess the degree to which a community will be resilient to a disruption. These indices often feature a hierarchical design, where several indicators are grouped into domains and subsequently aggregated into a single composite score or index. Despite their widespread use in research and practice, a clear consensus on validating these scores is lacking, which hinders an evidence-based measurement design and impacts the credibility of frameworks. As a result, notable inconsistencies exist in the selection and organization of indicators and domains, as well as in the methods employed to derive composite scores, such as various normalization and dimension-reduction techniques. This study evaluates the quality of existing community resilience scores and offers recommendations for enhancement. Previous studies have often concentrated on the initial impacts of natural hazards, validating scores with concurrent or convergent variables such as damage and casualties. In contrast, this study adopts a criterion validity approach, examining how well a framework and its derived score predict or correlate with specific outcomes essential to the concept of resilience. This involves establishing a benchmark for community resilience based on longitudinal community outcomes regarding population, employment, and life expectancy before and after disaster events, which are then transformed to quantify two core dimensions of resilience: the capacity to mitigate initial impacts and the trajectory of community recovery. Analyzing presidential disaster declarations over the last two decades in the conterminous United States, this study selected 1,716 counties that had one or more declarations, totaling $182 million in damages. Then, this study utilized Structural Equation Modeling (SEM) to evaluate several commonly-used resilience and vulnerability frameworks against these observed disturbances through the benchmark. The benchmark, in conjunction with SEM, serves as a quality assurance tool, providing empirical relationships between various levels of disruption, framework scores, and resilience outcomes. The result revealed that complex framework designs did not necessarily guarantee a better representation of community resilience. The hierarchical framework design presented challenges in criterion validity, particularly concerning the unintended and non-intuitive connections with the benchmark. These findings provide evidence-based recommendations for all stages of framework development and contribute to the measurement science of community resilience.

Pub Type

Talks

Download Paper

Local Download

Keywords

Community Resilience, Criterion Validity, Disaster, Indicator

Citation

Gu, D. , Dillard, M. and Gerst, M. (2025), Assessment of community resilience frameworks: a validation through community outcomes after disaster events, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=961187 (Accessed July 31, 2026)

Additional citation formats