Dynamics of hurricane risk perception

Within the field of disaster research, hurricanes present among the most challenging topics of investigation because of their unpredictability, regular occurrence, large scale, and multifaceted destructiveness. Despite the considerable research that has been conducted on the human and social dynamics surrounding hurricanes, there have been surprisingly few examinations of individual hurricane risk perception or optimistic bias. This paper presents an analysis of our investigation into hurricane risk currently underway.

In January 2006, a mail survey was sent to households in 41 counties adjacent to the Gulf of Mexico, excluding the areas affected by Hurricanes Katrina and Rita (62% response rate, n = 824). A follow-up survey was sent in January 2008 (52% response rate, n = 361). Measures included Peacock et al.'s index of hurricane outlook, evacuation optimistic bias, dispositional optimism, and hurricane experience (among others). Paired sample t-tests (n = 361) were used to compare individual scores across the two-year span of panel data collection.

Hurricane outlook, estimation of the average probability of a forced evacuation in the 2008 season, and estimation of the personal probability of a forced evacuation in the 2008 season all became more optimistic. There was no significant change in dispositional optimism. Results also show that perceiving less hurricane risk is associated with general optimism and being male, older age, and higher income. Many of these findings are consistent with risk studies involving other hazards. Having less past experience with hurricanes was also associated with perception of less hurricane risk, suggesting that the growing population of coastal residents who have little or no hurricane experience may underestimate the risk involved. It was also shown that perception of hurricane risk decreased following two mild hurricane seasons. Finally, both components of the optimistic bias measure became more optimistic: believing that one is less likely than others to suffer a major hurricane adjusted positively along with the increased optimism for others.

The results of this study are being used as a basis for a wider investigation into the function of information and social influence in hurricane evacuation orientation and behavior. In the present three-year panel study, survey and interview data will be collected from individuals located along the U.S. coast from Wilmington, NC to Brownsville, TX. The study includes development of a cognitive-affective model of hurricane risk perception, further investigation of optimistic bias, and real-time observation of hurricane evacuation decision-making. The first survey data were collected in June 2010. This presentation will also include preliminary results from the current investigation.