Abstract
Community members who evacuate to shelters may represent the most socially and economically vulnerable group within a hurricane’s affected geographic area. Disaster research has established associations between socioeconomic conditions and adverse effects, but data are overwhelmingly collected retrospectively on large populations and lack further explication. As Hurricane Florence approached North Carolina in September 2018, RTI International developed a pilot survey for American Red Cross evacuation shelter clients. Two instruments, an interviewer-led paper questionnaire and a short message service (SMS text) questionnaire, were tested. A total of 200 evacuees completed the paper survey, but only 34 participated in the SMS text portion of the study. Data confirmed that the sample represented very marginalized coastline residents: 60 percent were unemployed, 70 percent had no family or friends to stay with during evacuation, 65 percent could not afford to evacuate to another location, 36 percent needed medicine/medical care, and 11 percent were homeless. Although 19 percent of participants had a history of evacuating for prior hurricanes/disasters and 14 percent had previously utilized shelters, we observed few associations between previous experiences and current evacuation resources, behaviors, or opinions about safety. This study demonstrates that, for vulnerable populations exposed to storms of increasing intensity and frequency, traditional survey research methods are best employed to learn about their experiences and needs.