Proof that COVID-19 Data Accuracy depends on countries' development levels
If COVID-19 data accuracy did not depend on a country's development level, then there would be no correlation between the number of confirmed coronavirus cases per 100,000 people and GDP per capita, the Human Freedom Index, or the Corruption Perceptions Index.
A correlation coefficient values range from -1 to 1 and indicate relationships between two variables, where -1 is the absolute negative correlation, 0 is the absence of correlation, and 1 is the absolute correlation.
There is a strong statistically significant correlation between countries' confirmed COVID-19 cases and GDP per capita. The correlation coefficient is 0.8 (95% CI: 0.74 - 0.85), which wouldn't have existed if the data accuracy was the same in every country.
The chart includes only countries with a population of more than 300,000 people. Otherwise small countries like the Vatican would distort the values of confirmed cases per 100,000 people.
But even for the sample of all 197 countries, the correlation coefficient is 0.49 (95% CI: 0.38 - 0.59) that is still a strong statistically significant correlation.
For comparison, here's how the graph of coronavirus cases per 100K by total population looks like:
The correlation coefficient is -0.04 (95% CI: -0.19 - +0.11) and the regression line is almost parallel to the abscissa axis, showing no correlation, as expected.
173 countries with a population more than 300,000 ranked by confirmed COVID-19 cases:
Okay, but what does it all mean?
It shows that the number of confirmed coronavirus cases in each country is literally what the country has managed to detect.
But little it says about the actual numbers, especially in developing countries, as testing policies and capabilities vary greatly.
- It would make little sense to blame countries solely for being at the top of the infected list, as it's like blaming for being transparent and capable.
- Keep in mind while comparing that accuracy of the data depends on the level of a country's development.