Spielman Reproduction
In our reproduction study, we introduced a modification to Spielman et al.’s methodology by weighting the SoVI based on the percentage variance explained by each component in Principal Components Analysis (PCA). Inspired by Spielman’s suggestion that “components are sometimes weighted by their eigenvectors,” we assigned weights to the principal components, creating a weighted SoVI model for each of the 21 levels of analysis outlined in Spielman et al.’s original study.
The introduction of component weighting aimed to discern whether assigning different weights to components based on their contributions to overall variance would enhance the SoVI model’s internal and theoretical consistency. The percentages of variance explained were utilized as weights for the principal components.
Despite the innovative weighting scheme, our analysis, particularly illustrated through the California example, brought forth a crucial revelation. The ranked SoVI scores of counties exhibited inconsistency across models with different spatial extents, even after incorporating component weights based on the percentage of variance explained. This outcome underscores the intricate nature of social vulnerability assessments and the challenges in achieving spatial consistency.
It’s noteworthy that Spielman himself acknowledged the practice of weighing components in the construction of SoVI-like indices. However, our study adds a layer of complexity by integrating this concept within the framework of PCA and evaluating its impact on the SoVI model’s performance.
While the introduction of component weighting brought nuanced insights, the persisting spatial inconsistencies highlight the multifaceted challenges in social vulnerability assessments. As researchers explore innovative modifications, transparency about methodologies and their implications becomes increasingly paramount. Our study not only contributes to the ongoing discourse but underscores the intricate nature of SoVI models and the need for continued exploration and refinement in their application.
You can find my reproduction HERE
You can find the Spielman et al. repository HERE
Sources 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
Cutter, S. L., Boruff, B. J., & Shirley, W. L. (2003). Social vulnerability to environmental hazards. Social Science Quarterly, 84(2), 242–261. https://doi.org/10.1111/1540-6237.8402002
__