Reproducing Chakraborty (2021)

Despite achieving the same qualitative results of the original study (that PwDs are more likely to experience negative impacts from Covid-19 due to other demographic factors), the reproduction methods and exact results still deviated from the original paper. The lack of detail about the study’s methods caused slight operational changes to the data. Bigger deviations include changing from a Pearson’s to a Spearman’s correlation to test the relationship between variables and verifying the Kuldorff spatial scan statistic by generating clusters with SatScan and Spatial Epi.

I made edits to the original study to improve the visualization of the data and results. First, I added numbers to the figures to help the reader refer to them from the discussion. This reproduced study has more figures than the original study, so it is necessary to label the figures in order to refer to them. Then I added an additional visualization of a demographic factor of Census ACS data.

To improve the study and increase confidence in the numerical values assigned to counties, adjustments could be made to the clustering algorithm. The ability to reproduce Kuldorff spatial scan statistic was limited due poor control of the parameters used, so future reanalysis should employ an alternative clustering method such as gini-coefficient. Clusters with fewer counties were weighted more than clusters with more counties, so it would be valuable to find a statistical method that wouldn’t do this.

Related Links:

Link to reanalysis of Chakraborty (2021): https://eliseylchan.github.io/RPr-Chakraborty-2021/report/01-RPr-Chakraborty.html

The GitHub repository for this project: https://github.com/eliseylchan/RPr-Chakraborty-2021.git

The original study:

Chakraborty, J. 2021. Social inequities in the distribution of COVID-19: An intra-categorical analysis of people with disabilities in the U.S. Disability and Health Journal 14 (1):101007. DOI: 10.1016/j.dhjo.2020.101007.