Preparation for and response to rapid onset physical events such as earthquakes, tornadoes, and hurricanes test the ability of social systems to provide and meet the survival needs of a human population. This research will examine household pre-impact preparedness and post-impact restoration activities in South Florida following Hurricane Andrew. The central focus will be an examination of the ethnic differences in household response to warning and subsequent preparations as well as specific restorative activities undertaken following impact. Household preparation activities refer to potential loss of life and damage to property. Household restoration activities include those actions undertaken by households to deal with the emergency situation following impact and short-term actions undertaken to restore some semblance of normal household function. The multi-ethnic nature of South Florida will allow for a more complete examination and assessment of ethnic variation in these activities. This research will combine a field survey of approximately 300 households located in heavily impacted areas and telephone surveys of approximately 1000 households serviced by Miami media markets and in the hurricane warning zone to assess ethnic variations in preparedness and restoration activities. While the literature has often alluded to ethnic variations in disaster activities, the nature of these variations has yet to be fully research. Social science literature is replete with discussions of the importance of race and ethnicity for employment, income, education, socioeconomic status, housing quality and ownership, access to political power, family organization and structure, and the like. All of these factors play critical roles in preparedness and restoration activities. In addtion, there are often language and other cultural differences among ethnic populations, such as those of South Florida. From both a theoretical and applied perspective, it is important to understand the role of ethnicity in disaster response and management.