rozierguillaume / covid-19

How does COVID-19 spread around the world? Charts, data, dashboards and interpretations...
MIT License
307 stars 35 forks source link

Improving interpretability #4

Open nikohansen opened 4 years ago

nikohansen commented 4 years ago

Hello, very nice project! I very much appreciate that most data are shown in log scale (at least on the French data part)! One important issue remains to me: instead of showing absolute counts, showing the count per inhabitants (on the log scale) would improve the usability a lot. Absolute counts are close to meaningless unless one knows the population size from which they were taken. Per capita numbers don't really have any such drawback. Even for the interpretation of changes, the per-capita base line is somewhat relevant. The only situation where I would be interested to verify the absolute count is when the number is very small, like considerably less than 100. In any case, hover text could contain the absolute number to cross check.

The common normalization is per 100000 capita, AFAIK, whereas ourworldindata has chosen per one million.