Assessment of social vulnerability to natural disasters: a comparative study

Abstract

The purpose of this study is to examine and compare the methodologies being developed in assessing social vulnerability to natural disasters. Existing vulnerability literature shows that two methods have been used in developing social vulnerability indexes: (1) a deductive approach based on a theoretical understanding of relationships and (2) an inductive approach based on statistical relationships (Adger et al. in New indicators of vulnerability and adaptive capacity. Tyndall Centre for Climate Change Research, Norwich, 2004). Two techniques were also utilized in aggregating social vulnerability indicators: (1) a deductive approach using standardization techniques such as z scores or linear scaling (Wu et al. in Clim Res 22:255-270, 2002; Chakraborty et al. in Nat Hazards Rev 6(1):23-33, 2005) and (2) an inductive approach using data-reduction techniques such as factor analysis (Clark et al. in Mitig Adapt Strateg Glob Change 3(1):59-82, 1998; Cutter et al. Soc Sci Quart 84(2):242-261, 2003). This study empirically compares deductive and inductive index development and indicator aggregation methods in assessing social vulnerability to natural disasters in the Gulf of Mexico and Atlantic coastal areas. The aggregated social vulnerability index is used to examine a relationship with disaster losses in the Gulf of Mexico and Atlantic coastal areas. The results show that coastal counties with more vulnerability in terms of social achieved status are positively associated with disaster damages, while variations in the development of the index using deductive and inductive measurement approaches produce different outcomes.