Results found: 1

Search results

Search:
in the keywords: Benford's Law

Sort By:

Limit search:

The perils of premature evaluation: reassessing the application of Benford's Law to the USA's COVID-19 data

100%

Dutta-Powell R.

Statistics in Transition new series

2025

vol. 26

issue 3

1-13

This paper reviews earlier applications of Benford's Law to the COVID-19 data in the United States that claimed these data's non-conformity with Benford's Law, and uses later and more granular data to demonstrate that this was likely due to the earlier data being unsuitable for such applications. It also demonstrates that the same dataset, analyzed in different ways, can show vastly different levels of conformity with Benford's Law. Specifically, most US states show high degrees of conformity for the COVID-19 cases and cumulative deaths when the Robust Order of Magnitude (ROM) is over 3 and data at the county level is used to analyze state outcomes. Conversely, when the county data is aggregated to the state level and analyzed (i.e. case totals for all counties are summed to create a single state figure for each day of the pandemic), every state shows non-conformity. Only new deaths showed the reverse pattern - this is likely because new deaths at the county level do not span sufficient orders of magnitude, and aggregation to the state level overcomes this. This suggests that some instances of non-conformity with Benford's Law in the literature may be caused by its applications to inappropriate datasets or methodological issues.

Refine search results

1 Statistics in Transition new series

1 Dutta-Powell R.

1 2025

Search results

The perils of premature evaluation: reassessing the application of Benford's Law to the USA's COVID-19 data