Is storm surge scary? The influence of hazard, impact, and fear-based messages and individual differences on responses to hurricane risks in the USA

Abstract

This article uses data from an online survey-based experiment to investigate how risk communications and individual differences influence people's responses to approaching hurricane risks. Survey data were collected from 1716 residents of coastal areas of the USA affected by Hurricane Sandy. Respondents were randomly assigned to receive a combination of textual messages about a hypothetical approaching hurricane, including hazard-based, impact-based, and fear-based messages. The analysis examines how the experimental messages influenced respondents’ evacuation intentions, risk perceptions, efficacy beliefs, and perceptions of the information and its source. The influence of non-message factors, including respondents’ actual and perceived geographical exposure to hurricane-related risks, evacuation planning, and hurricane-related experiences, is also investigated. The results indicate that the high-impact and fear messages increased evacuation intentions, risk perceptions, and response efficacy, but the effects were small. The hazard message manipulations did not significantly influence most of the dependent variables examined; in particular, neither of the two storm surge messages tested increased evacuation intentions or risk perceptions relative to the wind-only or flood message. There were also no significant differences in message effects among respondents who lived or thought they lived in areas at higher risk. Further, several individual difference variables examined influenced evacuation intentions more than the message variations. Overall, experience evacuating for Sandy was the strongest predictor of evacuation intentions. These results indicate the importance of designing and evaluating hazard risk communications in the context of the other messages people are receiving and the individual differences that influence protective decision making.