摘要: |
Revealed preference is the traditional method to collect hurricane evacuation behavior data. However, revealed preference surveys, as they are currently administered, have the disadvantage that they are unable to collect time-sensitive and policy-sensitive data needed to test evacuation policies. Since time-sensitive and policy-sensitive data is necessary for effective evacuation demand modeling and no methods currently exist to collect such data, this study reports on the development and testing of a candidate procedure to address this need. The procedure involves using the stated choice approach to data collection adapted to collect dynamic information and enhance the realism of each scenario by presenting it in audio-visual form on a DVD. Nine hypothetical storms were presented in audio-visual form through a series of time-dependent scenarios to a random sample of respondents in the New Orleans metropolitan area. The new method was evaluated by collecting data using both new and traditional methods and comparing their cost and their ability to produce good evacuation models. In the new method, survey respondents watched animations of storm scenarios and stated how they believed they would behave in each time interval as the storm approached, while in the traditional method they reported on their actual behavior during Hurricane Gustav, which made landfall near New Orleans in 2008. Results indicate that the new stated-choice method is easy to use and effective in collecting time-dependent and policy-sensitive data but costs 25 percent more than the traditional method. The new method appears to have the potential of evolving into a survey instrument that can be used by researchers and practitioners working in hurricane evacuation modeling. |