search

daenuprobst / dashcoch

Interactive Web Visualizations of COVID19 Cases in Switzerland
20 stars 20 forks source link

Turn dot charts into bubble charts #6

Open Eisit opened 4 years ago

Eisit commented 4 years ago

The dot charts could have the size of the dot proportional to the Cantons' population.

See a US example here: https://www.axios.com/americas-coronavirus-cities-outbreak-5f22e933-42f1-464e-a68d-5f8ee0923fda.html

Eisit commented 4 years ago

The implementation is good. But it raises another question (which I don't know with certitude the answer to): should the trend line calculation weight each canton by number of cases? I'd say: -No for the prevalence vs. density graph, because prevalence already depends on nb of cases; -Yes for the CFR vs. pop.>65 graph, because the more cases there are, the more likely to be an outlier.