Effects of Hurricane Katrina on Evacuation Intent: A Panel Study

Florida International University, Miami FL

Investigators

Abstract

In this panel study a geo-coded sample of householders in Florida, Alabama, Mississippi and Louisiana who were first interviewed after Hurricane Ivan will be re-interviewed related to their hurricane knowledge, attitudes, behavior and future intent. The purpose is to assess the effects of different types or levels of experience with the 2005 hurricanes (including the "virtual" experience of watching the evacuations and hurricane impacts on television) on the attitudes and behavior, particularly those related to evacuation. Specific objectives include: 1) To document the extent to which each household experienced and responded to Hurricanes Katrina and/or Rita; 2) To determine how these households assessed their level of risk, including the information sources used; 3) To analyze household decision processes related to preparing and/or evacuating, including choice of routes. 4) To identify factors that promoted and constrained evacuation decisions; 5) To compare hurricane-related knowledge and attitudes before and after the 2005 hurricanes; 6) To analyze the relationship between the nature of their 2005 experiences and any revealed changes in attitudes, including future intentions; and 7) To develop a set of recommendations to policymakers and responders to improve hurricane response, including evacuation compliance, particularly among those at highest risk. An experienced interdisciplinary team of social scientists and a traffic engineer is designing and implementing the study, and interpreting the results. The research focuses on four areas which are essential considerations in any comprehensive model of evacuation behavior: Hurricane message and risk communication; time issues and decision constraints; transportation constraints; and high-risk populations. The major part of the data collection consists of Computer Assisted Telephone Interviewing (in English or Spanish) of as many of the original GIS-based random sample of 3200 households as possible. The Co-PI's have extensive experience in hurricane behavioral research, including evacuation decision-modeling, family and household response, and GIS-based data analysis. Panel studies are rare in disaster research. This project will provide unique information about the effects of subsequent hurricanes on the attitudes and behavior of the same respondents. In addition to their reported experience, the GIS-based data set will allow the researchers to geographically locate each household in relation to subsequent storms, including an analysis of reported evacuation routes. Another factor adding to the intellectual merit of this work is that the decision-making model refined using these data will add to a better understanding of cognitive processes related to how people define and react to risk. The findings from this project will be widely disseminated to researchers and practitioners through papers and publications, as well as posting of the report and data on a website. The work will result in recommendations to improve evacuation compliance, and thus promote the safety of coastal populations, including those most at risk.