Modeling Hurricane Evacutaion Decisions with Ethnographic Methods

This paper directly models individual and household hurricane evacuation behavior using ethnographic decision tree analysis. This approach uses a set of iterative processes to inductively derive a general decision model from specific individual decision models. To elicit the model described here, below the authors and several graduate students interviewed Miami residents who had been in South Florida during both Hurricanes Andrew in 1992 and Erin in 1995. The resulting model of hurricane evacuation decision processes was then tested with interview data collected from a separate random sample of 954 South Florida residents drawn from areas that were evacuation zones and areas immediately adjacent to them at the time of Hurricane Andrew. The model captures the complexity and messiness of real-life decision-making by including criteria-showing how people are constrained by their perceptions of the hurricane, the safety features of their homes, the time they have available to prepare for the hurricane, their age, and the reactions of other family members who are also deciding whether or not to evacuate. By showing the richness of the decision process as well as its messiness, results taken from this model can better inform emergency managers who need to know how people will react to the approach of a hurricane.