I just happened onto this very current article about mapping the spread of the coronavirus 19 and Mapping coronavirus, responsibly.
This article is talking about some of the kinds of distortions that mapping software can produce because we are mapping the nearly spherical earth onto a flat page. In particular, it's talking about how one's intuitions are getting influenced by these and other changes. Below is a wild example. What would you make out of a "chart" like this? How would this influence your intuitions your thinking about the risk of catching the virus? Would something like this have an impact on policy?
Have a read of the article to get a better understanding of using map charts responsibly.
Below is an example of an irresponsible map.
I'm reading "Good Charts" at the moment so this fit right in.This map shows nothing more and is more obstructive than the bar chart. This map is a graphic but is not informative at all.
The initial choropleth map also unfairly describes the risk of such neighbouring provinces like Hunan.
Proportional symbol maps are good though I worry about logarithmic scale graphs as I find some people have a difficult time grasping the concept of Log
This is a great read, thank you for sharing!I think that the blue chloropleth - the one adjusted to show to the scale - is actually significantly easier to understand at first glance than the bar chart. Many readers won't spend much more time with a graph than the initial first glance, and I think that's an important consideration.
The one that you pulled out here is wild and even as someone pretty skilled at reading weird visualizations, I don't really understand what it's trying to tell me.
Thanks for sharing Tom. This is an excellent very informative read on how to use mapping. The issue of using totals is something I might not have realized before reading this.
Maery Simmons (Past Member)
I do agree that I choose the most click bate graphic of all of the items in the article. And yes, it has no scale, and is all sorts of BAD.
But it is a very good article. I'm glad you enjoyed it.
--Tom
My peers are partial to this map:
https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
I do find this easier to find outlier for visual people who also may not know their geography. :) Also, wondering from my experience in mapping items if they are mapping lat/long or by zip. Most tools I've used for mapping omit any location with a population under 10K. (How rude for us rural folks).
I am not bothered by the visual representations with most mapping vizs with the exception of that awful one above! It's like the child of a bar chart and a heatmap. :)
I'm curious about that map ... it seems in some instances that the dots are city/province centered (China, USA, Australia) and in others (esp, Russia, Brazil, UK) it's country centered. The list on the left is by Country though
Likely has to do with the level of aggregation that various jurisdictions make available. (ETL always a drag and makes an impact on what you can report.)
And of course it's always fun to compare Scranton to Russia
But less flippantly and more on topic that info panel on the bottom (which has the point level info) is incredibly an important part of this dashboard. I try to include Text info descriptions on my dashboards.
If you want to try to do your own Maps. The following data might be interesting.
https://github.com/CSSEGISandData/COVID-19
This is the map that we have been using here too. What's interesting to me is that it has changed recently for the US - before it was showing cases by city/county and it has moved to show by state. I do think that not having any relation to population density really skews the information.