The study is a contextual analysis of flood victimization. Victimization is defined as the social, psychological, and physiological aftermath experienced by victims of a disaster. Disaster researchers concentrate on the victims' characteristics to explain the varying degrees of their victimization, providing only ambiguous results. Theorists such as Kreps, Wildavsky, and Douglas contend that the outcomes of disasters are contingent upon social structure. This analysis treats victimization as one such outcome. The condition and behavior of individuals can be explained by the presence of disaster and the conditions of social organization. A model explains victimization based on individual's attributes (individual resources), his social environment, and the disaster characteristics. This study uses the 1984 Mingo Creek Flood Victims Survey data to test the model. The data contain information measuring victimization. The survey data are linked with 1980 Census tract data. The tract data provide indicators of the social networks. This tract information, the contextual variables, taps the social conditions, including poverty, unemployment, geographic mobility, and family patterns. This study uses factor analysis to identify the dimensions of victimization. Regression tests the relationship between the contextual variables, the individual resource variables, the disaster characteristic variables, and victimization. The results of the analysis show that victimization is multidimensional with different types of variables being significant predictors for each dimension of victimization, one variable indicating the intensity of the disaster, the dollar value of damage victims experienced, is found to be a significant predictor of the psychological, physiological, and social disruption aspects of victimization. Variables measuring the family and unemployment patterns in the victims' census tract are significant predictors of the psychological and social disruption aspect of victimization. The findings provide general support for the proposed model of victimization. However, victimization is multidimensional with each dimension having a unique set of predictors. Based on the findings, this study suggests that future research focus on measurement and conceptualization of the characteristics of disasters and the victims' social environment.