Makeover Monday: WSJ Measles Vaccination Chart

July 10, 2017

Using ggplot2 and animation packages to reproduce and animate the famous Wall Street Journal Measles Incidence Chart - this time as a cartogram!

Last week I wrote a post about recreating the WSJ measles vaccination chart as an animated cartogram - you can check out the full write-up of the development here: www.mikelee.co/posts/2017-06-28-wsj-measles-vaccination-chart/

It was featured on DataCamp across all their social media platforms - pretty neat!

A comment from one of the readers (which disqus seems to have overwritten as I was re-initializing my site) read as so:

Your measles chart is a beautiful piece of work! No question. The (vertical) sorting merely by alphabetic state listing, however, adds nothing to the information. Even better, I think, would be to place these stats on a US map. The animation would then show the geographic progress of the disease. Try first on a state basis; counties would be too confusing The ultimate presentation, however, would show the importance of high population densities (metropolitan areas). - Kenneth Holland.

That’s an interesting idea Kenneth!

I wonder whether a similar type of graphic representation was considered by the WSJ authors. Either way, this is a great excuse to jump on the Makeover Monday train.

Makeover the WSJ Measles Incidence Visualization as a Cartogram

Given that I’ve already completed the data acquisition and manipulation in my previous post, I’ll jump right into plotting the state level data. The county-level piece is an interesting idea, but I’ll leave that for someone else!

Here is the final result:

Measles Animation Cartogram

Measles Animation Cartogram

I customized the plot in a few ways - first, I chose to represent the change in time via a timeline above the cartogram. Other developers have represented change in time using years as text - such as the Financial Times’ income distribution gif (their original post here)

I selected a moving dot for a few reasons:

  • given the large selection of years, a frequently changing portion of the image would be a distraction
  • the movement of the dot is an intuitive representation of change over time
  • the point could be encoded with data - the color represents the national mean incidence!

The final product is a quality alternative to the WSJ’s original visualization, though I prefer the original representation as a full grid over years - the drastic reduction in measles incidence is more impactful with 30+ years of negligible incidence stacked up next to each other. Have other thoughts? Leave them in the comments - hopefully disqus doesn’t delete em this time :p


Interested in learning more? Hire me to consult on your next project, follow me on twitter, leave a comment, or contact me via email. All inquiries welcome!