Social vulnerability indices are often used in studies of natural hazards to measure social dimensions of vulnerability. However, it is challenging to create a single index that encompasses all hazards, because indicators vary with the hazard context. Index modelers have tended to adopt quantitative, non-context specific approaches to this problem by selecting indicators primarily based on their use in previous studies. Qualitative disaster case studies, however, offer rich, detailed, and context-specific understandings of social vulnerability that are often absent from modeling studies. How would a social vulnerability index based on the indicators selected by modelers differ from one with indicators selected from disaster case studies for flooding? This study developed and analyzed hierarchical social vulnerability indices, using a study area of St. Louis, MO due to its high susceptibility to flooding hazards. Indicators drawn from both the modeling and case study research were divided into subthemes such as socioeconomic status, demographic structure, housing, or special needs and aggregated to build the index. The frequencies of indicators in each group were used to assign weights to each indicator. The resulting social vulnerability indices were mapped to show which areas of St. Louis are more socially vulnerable.