Objective: Data from a sample of 391 respondents who responded to ABC News Hurricane Katrina Anniversary Poll, August 2006, are used to test the relationship between predictor variables (social vulnerability index, insurance and severity of damage in living area) and the dependent variable (perceived disaster impact). Methods: Multiple regression is used to test the theoretical relationships. The dependent variable disaster impact is regressed on seven predictor variables simultaneously. Results: Social vulnerability has a statistically significant, positive effect on perceived disaster impact. Fully insured variable has a statistically significant, negative effect on disaster impact. Finally, severely damaged living area has a statistically significant, positive effect on disaster impact. Conclusion: The social vulnerability theory holds true in the context of Hurricane Katrina. Those who possess mixed characteristics of social vulnerability tend to be more affected by the impact of a disaster. Insurance tends to help lessen the disaster impact when individuals have the policies that cover all of their property. Finally, those who live in areas that are severely damaged from a disaster tend to experience more impact than those whose living areas are less damaged or not damaged.