Recovering well: assessing the Social Vulnerability Index's ability to account for household recovery outcomes after the 2015 Gorkha Earthquake

Abstract

This study investigates to what extent the Social Vulnerability Index (SoVI) can account for household recovery outcomes after the 2015 Gorkha Earthquake in Nepal, using household survey data collected in 2019 from some of the most hit districts in the country. The results of this study found across the different recovery outcomes, there were different SoVI subcomponents which had statistically significant influence over the recovery outcomes in different directions and with different magnitudes. SoVI subcomponents did not always recovery outcomes in expected ways, with some subcomponents reflecting higher vulnerability scores associated with better recovery outcomes. For example, Migration & Gender vulnerability was found to be consistently associated with better household recovery outcomes in housing stability, short-term economic stability, physical health and earthquake-related trauma symptoms. Even though Total SoVI had a statistically significant relationship with 7 out of 11 selected recovery outcomes, using the SoVI subcomponents as predictor variables instead produced models better fitted to the data for almost all of the recovery outcomes. These results reveal that the current non-weighted Total SoVI scores are unable to capture the unequal effects of the SoVI subcomponents on different recovery outcomes. Instead, a weighted SoVI model, tailored to the specific recovery outcome and the time after the disaster being predicted, is able to provide a more accurate account of recovery outcomes. More accurate models to predict specific recovery outcomes using SoVI subcomponents can inform the distribution of aid in an equitable manner that targets socially vulnerable communities that need it the most.