Hidden Common Factors in Disaster Loss Statistics: A Case Study Analyzing the Data of Nepal

抄録

This study aims to examine common hidden factors in disaster loss statistics and identify clues for verifying the fitness of the global targets of the Sendai Framework for Disaster Risk Reduction 2015–2030 (SFDRR) to rule countries’ effort in reducing disaster risks. In this study, we first conducted an exploratory factor analysis (EFA), followed by a confirmatory factor analysis (CFA) using structural equation modeling (SEM). As a result of the EFA, we were able to extract three factors, namely Housing, Casualties or Education, and Relocation. In the analysis of SEM, we assumed three latent variables based on the results of the EFA. The relationship between the latent and observed variables was established in a manner that conformed to the implications of the EFA. According to the SEM results, we eventually identified three latent variables, namely Housing, Education and Relocation, as hidden common factors. Based on this identification, our judgment indicates that the latent variables appeared to be related to the following global targets of SFDRR: (b) those concerning the number of affected people and (d) those concerning damages to infrastructure and disruptions to basic services. It was found that relationships between variables could be clearly illustrated by using the path diagram. This study can be considered as a good example of introducing SEM to visualize hidden common factors and their relationships in an intelligible manner. Based on the results, we propose a starting point for discussing the fitness of SFDRR’s global targets by utilizing EFA and CFA (SEM) techniques. The path diagram can indicate the extent to which the indicators contribute to global targets that will be represented as latent variables. In the end, explicit reference should be made to the material data’s limitations in the disaster loss statistics. An effort to elaborate the input data themselves must be made in the near future.