A multilevel analysis of public risk perceptions of climate and weather

Public risk perceptions regarding climate and weather are known to be influenced by demographic factors including direct personal experiences with weather events. These perceptions are dynamic, often shaped by the immediacy and visual impact of weather phenomena and climate-related disasters, alongside evolving scientific understanding and projections of future climatic trends and their potential impacts on the environment and human society. In this paper, we present a multilevel analysis of public risk perceptions on climate and weather, leveraging the 2021 wave of the Lloyd’s Register Foundation World Risk Poll (WRP), a large survey dataset of 129,992 individuals interviewed across 121 countries. Our project aim is to test whether individual-level factors such as age, gender, education, household income, household size, whether their living environment is urban or rural, and personal experiences of severe weather events shape public risk perceptions on climate and weather. In this study, we also integrate several country level indices into the WRP dataset to generate insight on the relative influence of individual-level versus national-level factors in influencing our outcomes, including EM-DAT data on natural disasters (the number of climatological, geophysical, hydrological, and meteorological disasters in each country), ND-GAIN data on projected changes in hazards (future projections for heatwaves, flooding, and sea-level rises in each country), and WGI data on governance in each country. Using multilevel modelling techniques, we provide empirical validation of the effects of a list of factors that are theorised to shape climate and weather-based risk perceptions. The addition of random slopes (i.e. the intercepts is the same for all countries, but the slopes vary for each nation state) to our multilevel models also allows us to examine how the nature of the relationship of the identified factors and our outcomes varies by national context. The current findings highlight the dynamics between individual factors and larger environmental and governance contexts in shaping public risk perceptions of climate and weather. They form the basis for the future phases of our project development, which include the selection of several countries, representing different climate and social contexts, to use as case studies to test the effectiveness of early warning systems for disaster risk reduction. Our study and research findings offer important insights both for policymakers in crafting more effective, context-sensitive strategies for climate mitigation and adaption, and for national meteorological services in developing early warning systems for severe weather events.