Exploring key social capital indicators for disaster preparedness in rural disaster-prone areas: a boosted regression tree approach

Social capital provides a valuable theoretical framework for boosting disaster preparedness. However, the multidimensional and intricate nature of social capital poses challenges in its measurement. Achieving a balance between comprehensive and effective measurement of social capital indicators necessitates additional exploration, especially within the specific context of disasters. The current study utilizes the boosted regression tree (BRT) approach to identify key social capital indicators that influence disaster preparedness in rural disaster-prone areas of China. BRT is highly regarded for its ability to capture complex nonlinear relationships and interactions among variables, providing accurate predictions and facilitating interpretability for practical applications. Results reveal that geographic close social ties, social status, collective resources, non-farm employment assistance, gift exchange, interpersonal trust, and sense of belonging significantly impact disaster preparedness. The findings could offer valuable guidance to policymakers in designing targeted intervention strategies aimed at enhancing disaster resilience within these communities.