Social Vulnerability Index: Replicating Spielman et al. (2020)
In order to improve the replication of Spielman et al. (2020), I focused on creating figures and tables as well as reweighted the national SoVI model by percent variance of each component. The addition of a national SoVI map helps to showcase what areas of the country are driving the national model. For example, several of Alaska’s counties having some of the highest social vulnerabilities may connect to why the much of the continental US showcases medium to lower social vulnerability in comparison; this observation can help us understand the differences in social vulnerability models within California (Figure 1 in the original study). The reweighting of the national SoVI model using percent variance of each component is an important step in reevaluating what our replication of Spielman et al. (2020) should represent.
Particularly with the national SoVI map figure, this replication study emphasized the importance of visual ‘checks.’ Having snippets within the code that can visualize and present what is in the code is helpful for future readers and researchers when reasoning behind a choice or the impact of a choice is not clear. The reweighting adjustment of the SoVI model also highlighted the importance of revaluating variables and representations in a replication study. The importance of choosing the right measurements for what the replication wants to observe is crucial for the tranparency and validity of both replications and reproductions. These additions and edits to the replication of Spielman et al (2020) will improve how the original study evaluates social vulnerability across the United States.
Check out the replicated study here
Check out the Github repository for the replication here
I would like to specifically thank fellow students Grace Sokolow, Audrey Park, and Sydney Alexander for the collaboration on this replication.
References
Spielman, S. E., Tuccillo, J., Folch, D. C., Schweikert, A., Davies, R., Wood, N., & Tate, E. 2020. Evaluating social vulnerability indicators: criteria and their application to the Social Vulnerability Index. Natural Hazards, 100(1), 417–436. https://doi.org/10.1007/s11069-019-03820-z