How are various natural disasters cognitively represented? a psychometric study of natural disaster risk perception applying three-mode principal component analysis

This study explores the features and structure of laypeople’s risk perceptions of natural disasters using a psychometric paradigm (PP) that employs three-mode principal component analysis (3MPCA) in Japan, a country with high vulnerability to various natural disasters. Laypeople (n = 825) and natural disaster experts (n = 22) living in Japan answered a questionnaire on judgments of 11 risk characteristics (e.g., extent of dread, personal controllability, scientific knowledge) and four risk perception items (subjective risk assessment, need for government measures, need for individual mitigation, and risk acceptance) regarding nine natural disasters. 3MPCA revealed a three-mode dimension structure that consists of three scale components (dread, controllability, and unknown), three target components (localized catastrophic, drastic, and gradual disasters) that are interpreted as cognitive disaster types and seven person components (each dimension of dread or controllability according to the target components and common dimension of unknown of all hazards). Furthermore, the cognitive disaster types varied between laypeople and experts. Multiple regression analysis revealed that dread determined risk perception items, except for risk acceptance, which was determined by controllability. Importantly, the effect of risk characteristics judgment varies according to the cognitive disaster type. This result indicates that the structure of natural disaster risk perceptions differs according to people’s recognition of hazard properties. Therefore, 3MPCA is a useful method for exploring such a structure to obtain a deeper understanding of the nature of hazards.