The Antecedents of Risk Perception of Natural Hazards: A Meta-Analysis

To ensure better preparedness and response to natural hazards, involving people is crucial. Among the determinants of individuals’ actions to protect themselves, both before and after a natural hazard, is risk perception (van Valkengoed & Steg, 2019). A conspicuous number of studies investigated the antecedents of risk perception; nevertheless, there are no quantitative summaries (i.e., meta-analysis) of the available research. The aim of this contribution is twofold, namely, to estimate effect sizes of different potential antecedents of risk perception and to investigate the effect of possible moderators of such relationships. A query string was used on Scopus, Web of Science, and PsycINFO, and a final number of 121 records (127 studies) was included based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Results of the first aim showed that, among thirty-two predictors identified, nineteen reported significant effect sizes. They can be organized into three clusters (Bonaiuto & Ariccio, 2020): 1) factors related to the relationship individual-risk, 2) factors related to the relationship individual-community, and 3) individual factors (i.e., sociodemographic and dispositional factors). The first cluster showed the highest number of variables with strong effects. Overall, an effect of publication bias and study heterogeneity was observed. Regarding moderators, the type of natural hazard considered and the level of risk area of the sample proved to be relevant in some cases. Existing research is scanty regarding some natural hazards (e.g., landslides, drought); thus, future studies may consider studying risk perception antecedents in relation to these understudied natural hazards. Policymakers should consider the three clusters of antecedents. This represents a novel attempt to quantitatively summarize the available literature on the topic of the determinants of risk perception, considering interdisciplinary studies, including grey literature, and with no limits in terms of publication date.