There is currently a lot of confusion on the actual number of deaths caused by COVID-19. The Center of Disease Control and Prevention (CDC) currently reports two numbers.

One is the number of confirmed and probable deaths caused by COVID-19 and the other is the number of death certificates that list COVID-19 as the cause of death.

As of May 22nd, the total number of deaths due to COVID-19 are 97,049 and 73,639 using the first and second methods, respectively. When the difference between these two numbers became apparent to the public, many people started claiming that the number of deaths was greatly exaggerated.

Upon further investigation, the second number suffers from data lag due to the time it takes death records to be submitted to the National Center for Health Statistics (NCHS). The CDC reports that the data are lagged by an average of 1-2 weeks. The table shown on the CDC website shows the provisional deaths only go up to the week ending on May 16th, 2020. That must explain the huge discrepancy total death count, right?

To find out if this is actually true, we used the most trusted name in research to find the first 11 websites reporting total number of deaths on May 16th, 2020.

The average number of deaths reported by the eleven media sources on May 16th was 87,681 compared to the CDC’s reported number of 73,639. Figure 1 shows that no media source reported the number below the actual number retrieved from the NCHS database. The media appears to have over-reported the number of deaths by 17.4%.

To determine if the number being reported is significantly different, we’re going to use a one sample t-test where the null hypothesis is that the population mean is equal to 73,639 and the alternative hypothesis is that the population mean is not equal to 73,639. In other words, the null hypothesis is that the numbers reported are the same and the alternative hypothesis is that the numbers are different. We’re going to skip the math and just show the important numbers in Figure 2 below:

Given the calculations above, since *t-test* is larger than *t-critical*, we reject the null hypothesis and conclude that the number reported by the media is significantly different from the actual death count.

Disclaimer: The Harold Herald does not try to influence anyone’s thinking; in this article, we’ve merely pointed out some numbers and will allow each person to make their own conclusions.