I am interested in unpacking the data that go into the making of the Pandemic Vulnerability Index (PVI). The index is an aggregation of various indicators that are normalized and weighted as outlined here. The specific datasets (“components”) comprising the current PVI model were each assigned to an indicator (“data slice”) as part of four major domains: Infection Rate, Population Concentration, Intervention Measures, and Health & Environment. I will be looking at the source data and unpacking the components aggregated within the four major domains to better understand the choices made in creating the PVI using as a framework steps from the OECD's Handbook on Constructing Composite Indicators:
- Step 1: Theoretical framework
- Step 2: Data selection
- Step 3: Imputation of missing data
- Step 4: Multivariate analysis
- Step 5: Normalisation
- Step 6: Weighting
- Step 7: Aggregating indicators
- Step 8: Sensitivity analysis
- Step 9: Link to other measures
- Step 10: Visualisation
Mapping individual indicators accross the United States to get a sense of their geographic distribution.
For interest in {Deaths,Cases}, and indicators in the dataset, I calculated 2 sample t-tests on the following hypotheses:
H0: mean(interest | indicator abv average) = mean(interest | indicator bel average)
HA: mean(interest | indicator abv average) != mean(interest | indicator bel average)