search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.12k stars 18.4k forks source link

Don't we need the number of total tested? #1101

Open yonggeun opened 4 years ago

yonggeun commented 4 years ago

While I was checking the tables from several countries, I realize that the trends with confirmed, deceased, and recovered could be more meaningful with the daily and accumulated number of tested from each country. You can see the example. https://www.kaggle.com/kimjihoo/coronavirusdataset#Time.csv Based on this table the number of total testees in South Korea is over 0.5% from its population.

There is a list with the number of testees here, https://ourworldindata.org/coronavirus-testing-source-data

alexkychen commented 4 years ago

I agree. Including numbers of testing would be more informative than reporting only numbers of confirmed cases. For example, we could estimate how many infected cases exist in a population by the following simple equation.

Estimated infected cases = (no. of confirmed cases / no. of testing) * population size

I am not an epidemiology expert, but this is how we ecologists would do for wild populations. Of course, the equation is under the assumption that tested cases are randomly sampled in a population. However, I guess at this point most tested cases are skewed to infected populations.

atmacafurkan commented 4 years ago

Some countries report official numbers, but many sites online that I can find are far behind the current dates. It would be good if the number of tests is included.