Application of Benford’s Law to Detect if COVID-19 Data is under Reported or Manipulated

Abstract

Tackling of COVID-19 has been the prime concern for all the countries because of the virus’ deadly tenure across the globe. But an attempt to curb the virus requires an effective analysis of its spread. Falsified data is suspected by experts on account of policy-making, political strategies and fear handling. Benford’s law, which is an effective and widely used technique to detect fraud and fabrication, is employed and applied to data made available publicly, by the authorities, in an attempt to see if the data adheres to the Benford’s Law and if it is correctly reported or not. Benford’s law works well on exponential data and is compared to the initial growth data of the countries, but in case of control interventions an epidemic growth model is applied to the data and then goodness-of-fit test is conducted to see if Benford’s Law is satisfied for the selected three countries. Also, the level and rate of control is calculated to determine the deceleration in growth for each country.