Resilience measurement for environmental shocks and stressors: scale development and psychometric assessment for coastal urban informal settlements in Fiji and Indonesia.

Abstract

Background: Climate change and weather-related hazards, such as droughts and floods, pose substantial threats to the human health and well-being, especially for those in low-income households and informal settlements. Resilience, defined as the ability to cope, adapt, and recover, is critical for communities to manage these evolving threats. While there has been increased interest in ensuring that global public health and development programs contribute to resilience, the lack of valid and user-friendly resilience measurement tools limits the evidence base on the effectiveness of interventions to build resilience.

Methods: We developed scales to measure economic, environmental, and social resilience to environmental shocks/stressors among urban informal settlements in low- and middle-income countries. Using an evidence-based conceptual framework, we collected data from 882 households in coastal informal settlements in Makassar, Indonesia and Suva, Fiji. We used factor analysis and item response theory approaches and assessed internal scale validity, reliability, and measurement equivalence.

Results: Analysis supported a one-factor model for economic resilience, which showed a positive correlation with a financial satisfaction item, providing evidence of internal construct validity. The results also indicated a four-factor model for social resilience, with subscales for inclusion, social cohesion, collective efficacy and action, and preparedness. These subscales correlated with relevant external items-community satisfaction and perceived safety-supporting internal construct validity. The environmental resilience scale performed poorly in item response theory analysis and requires substantial refinement. The economic resilience scale demonstrated configural, metric, and scalar equivalence, suggesting that scores are comparable between households in Indonesia and Fiji. The social resilience scales showed only configural equivalence, indicating potential differences in how items relate to the underlying construct across countries. Both the economic and social resilience scales demonstrated acceptable reliability, with omega coefficients > 0.70.

Conclusions: We developed and internally validated scales to measure economic and social resilience to environmental disturbances that quantify resilience as a latent construct and are grounded in resilience theory. These scales are suitable for application in urban informal settlements in Indonesia and Fiji. We recommend their use, with re-validation as needed, in the monitoring and evaluation of resilience-building interventions and policies targeting urban households in low-income settings.