The influence of cultural worldviews on people’s responses to hurricane risks and threat information

Abstract

This article explores whether and how cultural worldviews influence people’s responses to approaching natural hazard threats, using data from a survey that asked residents of coastal Florida, USA, about a scenario of a hurricane forecasted to affect their region. The analysis finds that stronger individualist worldview is associated with lower hurricane-related evacuation intentions, cognitive risk perception, negative affect, response and self-efficacy, perceptions of government preparedness, and information seeking. It is also associated with greater hurricane information avoidance, perceptions of bias in key hurricane information sources, and negative reactance to hurricane threat information. Mediation analysis indicates that stronger individualists’ lower evacuation intentions are explained by their lower cognitive risk perception, negative affect, and response efficacy related to the hurricane threat. Their greater negative reactance to the hurricane threat information is partly explained by their lower cognitive risk perception, but not by their lower efficacy; both are counter to the predictions of the Extended Parallel Process Model. These results demonstrate that worldviews can interact with how people perceive and respond to near-term hazard risks and risk information, and they suggest that hazard risk communication should consider cultural theory when designing and conveying forecasts, warnings, and other risk messages. The authors recommend further research to investigate in greater depth the roles of cultural theory in these contexts.

Keywords:

• Hurricanes
• cultural theory
• worldviews
• hazards
• risk
• evacuation

Acknowledgements

The authors thank Jeffrey Lazo for his collaboration on the research discussed in the article, including coordinating the survey design and implementation.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The survey also measured respondents’ intentions to take other protective actions, but these variables exhibited limited variance and so are not examined here.

2 Additional details on the survey, including a map of the region sampled and graphics depicting the message stimuli, can be found in Morss et al. (2016).

3 The hurricane scenario and associated forecast products were developed for this study by the National Hurricane Center (NHC) and the NWS Miami-South Florida Weather Forecast Office.

4 All of the factor analyses discussed in this article used principal axis factoring and an oblique promax rotation.

5 Supporting this decision, Lazo et al. (2015) also found that this same item loaded onto both factors, and Lazrus (2016) found that its phrasing was confusing to some interviewees.

6 Median imputation is considered adequate when less than 5% of data are missing for a variable (Harrell 2001). Only age exceeded 5% missing data.

7 For example, the partial correlations between individualist worldview and each response measure control for egalitarian worldview and political ideology.

Additional information

Funding

This work was supported by the National Science Foundation under grants 0838702 (which was jointly funded by the National Oceanic and Atmospheric Administration) and 0729302. The National Center for Atmospheric Research is sponsored by the National Science Foundation.