A Mass of Covid-19 maps and data graphics

Thanks to John Scalzi (@scalzi), the prolific tweet author and semi-prolific sci-fi author, I just learned about a mapping and data dashboard for covid-19 from JHU: https://coronavirus.jhu.edu/us-map. Here’s my home of Cumberland County, Maine:

Image captured from JHU, data as of 15 April 2020

Image captured from JHU, data as of 15 April 2020

The default map is of absolute data (bad!) but the tabs at the bottom allow one to see standardized data (confirmed cases by population, fatality rate).

Click on the county to get the inset window; click on the chart in the inset to embiggen:

Screen Shot 2020-04-16 at 4.25.27 PM.png

Go play and look for trouble spots. Cumberland looks good in terms of the number of hospital beds, but that’s because it’s home to both Maine Medical Center in Portland and also Mid-Coast Hospital in Brunswick. A lot of medical cases are transferred to MaineMed from around the state.

Here’s a couple of images that nicely display the problem with mapping absolute data. Here’s the area around Madison, Wis., down to Chicago, in absolute values: Madison is the isolated red hotspot in the center:

Screen Shot 2020-04-16 at 4.53.04 PM.png

Now here’s the same area, but confirmed cases mapped by population, and Madison sinks into a general, medium-level of infection rate:

Screen Shot 2020-04-16 at 4.53.13 PM.png

I’d prefer greater clarity: legends for the maps to explain the color schemes; larger diagrams. It’s as if the whole site was designed for a large 27” desktop monitor, so rather crowded on my little 13” laptop screen.