An Assessment of Change in Risk Perception and Optimistic Bias for Hurricanes Among Gulf Coast Residents

• Published in: Risk analysis 2014-06, Vol.34 (6), p.1013-1024
• Main Authors: Trumbo, Craig, Meyer, Michelle A., Marlatt, Holly, Peek, Lori, Morrissey, Bridget
• Format: Article
• Language: English
• Subjects: Analysis; Bias; Biological and medical sciences; Coastal environments; Cyclones; Cyclonic Storms; Demographic variables; Demographics; Disasters; Earth, ocean, space; Education; Emergency preparedness; Exact sciences and technology; External geophysics; Gulf Coast; Gulf of Mexico; Gulfs; Humans; Hurricane Katrina; Hurricane Rita; Hurricanes; Income; Medical sciences; Meteorology; Optimism; optimistic bias; Order disorder; Perception; Perceptions; Public health. Hygiene-occupational medicine; Risk; Risk analysis; Risk Assessment; Risk perception; Storms, hurricanes, tornadoes, thunderstorms; Studies; Surveys
• ISSN: 0272-4332; 1539-6924
• DOI: 10.1111/risa.12149

This study focuses on levels of concern for hurricanes among individuals living along the Gulf Coast during the quiescent two‐year period following the exceptionally destructive 2005 hurricane season. A small study of risk perception and optimistic bias was conducted immediately following Hurricanes Katrina and Rita. Two years later, a follow‐up was done in which respondents were recontacted. This provided an opportunity to examine changes, and potential causal ordering, in risk perception and optimistic bias. The analysis uses 201 panel respondents who were matched across the two mail surveys. Measures included hurricane risk perception, optimistic bias for hurricane evacuation, past hurricane experience, and a small set of demographic variables (age, sex, income, and education). Paired t‐tests were used to compare scores across time. Hurricane risk perception declined and optimistic bias increased. Cross‐lagged correlations were used to test the potential causal ordering between risk perception and optimistic bias, with a weak effect suggesting the former affects the latter. Additional cross‐lagged analysis using structural equation modeling was used to look more closely at the components of optimistic bias (risk to self vs. risk to others). A significant and stronger potentially causal effect from risk perception to optimistic bias was found. Analysis of the experience and demographic variables’ effects on risk perception and optimistic bias, and their change, provided mixed results. The lessening of risk perception and increase in optimistic bias over the period of quiescence suggest that risk communicators and emergency managers should direct attention toward reversing these trends to increase disaster preparedness.