Compliance of case and death data published by country with benford law during the COVİD-19 pandemic

Recently, the increasing use of fraudulent or misleading data in statistical reporting in the scientific world has led those working in these fields to develop effective analytical tools to detect misleading activities. In this context, Benford's Law draws attention as a simple and powerful tool. The accuracy of case and death records published during the Covid-19 pandemic period has been the subject of debate all over the world. The aim of this study is to examine the compliance of published Covid-19 case and death data with Benford's Law. The data in the study were obtained from the data published by WHO for 119 countries in the first 3-month, 6-month and 12-month periods. Primary Care Test was used in the study. The conformity of the number of cases and deaths of the countries with Benford's Law was evaluated by Chi-Square, Euclidean Distance, Hotelling T, KS, Chebyshev Distance, Joenssens, Freedman-Watson-U2 tests. The concordance rate was found to be low in the number of cases and deaths; it was observed that the discordance was higher in the death data than in the case data. The findings of the study show that there is no significant relationship between case and death mismatch rates. In other words, it indicates that countries may be presenting case data correctly while not presenting death data correctly. This situation is similar to the development index of the countries. In other words, the high level of development of the countries does not ensure that the percentage of compliance is also high. When analysing such data, it would be appropriate to evaluate parameters such as the number of cases, deaths and tests at the same time, and to evaluate different tests together instead of a single test. It is useful to discuss the reasons for the incompatibility of the data in the light of many factors such as populist concerns of the governments of the countries, health policies, democracy indicators, diagnosis, record reliability and the course of the virus. Analysing the accuracy of these data with statistical methods will ensure public confidence in pandemic data, radical and effective measures and compliance with the measures.