Fascinating study of countries' reported COVID cases using Benford's Law. If you do not trust your country's reported number of cases/deaths, this is a statistical way to check their veracity.
Benford's Law says that leading digit of numbers from real data sets follow a certain distribution that favors small digits. Eg. the leading digit should be 1 about 30% of the time, 9 only 5% of the time.
When people make up numbers, they typically violate this distribution.
So @zekib collated data from many countries, and analyzed them using Benford's Law. You can see the results on page 27 of the following link.
app.powerbi.com/view?r=eyJrIjo…
The reported case numbers from Switzerland follow Benford's Law pretty well.
Case numbers from Russia, Turkey, and the US clearly diverge from what Benford would predict.
Keep in mind that the test is statistical. It needs large numbers, and it needs the data to span multiple orders of magnitude. Even then, a divergence doesn't immediately mean malfeasance; there may be other explanations.
Nevertheless, many people distrust official numbers, and this is a way to perform a sanity check.
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.