As a non-american I learnt a lot from plotting a timeline of US presidents.
I started this to practice wikipedia export parsing, I originally thought this would be an easy tasks as presidents of the U.S. are probably within the most documented thing in the internet. Although it proved more difficult than I expected I manage to get a workable solution.
The first step is to export the Wikipedia files from the US presidential category. Basically, this step will export a xml file with the wikipedia pages of the US presidents.
After battling a bit with text search and the xml2 package, the second
step is to parse the wikipedia data into a data frame. This become more
problematic than I thought because the standard for "terms" is a bit diffuse.
It easily gets mixed with Congress and Senate terms. As an example,
Cleveland
has to non-consecutive presidencies
(he is both the 22nd and the 24th president of the US if that makes any sense)
and it was impossible to distinguish that from a non-presidential term.
The most problematic bit though is the born place. I took a bit of a crazy
step and I decided to just capture the state where they were born using
a big regular expression. This approach yielded some errors I had to
correct manually.
With these data in a table I plotted both the timeline and the map of the United States. For the timeline I chose to represent their whole life from birth to death, highlighting their presidency time. I was surprised how consistent this was compared to some EU countries like Spain.
I also calculated the average assumed presidency time and death time which I then manually plotted using inkscape.
Finally I also gathered a collection of emoji from the openmoji site to use in the visualization.
After all of this plotting I started the manual curation of the items in inkscape. I basically moved around legends (and the islands of Hawaii) and added annotation text an arrows. I also added 3 points in the presidents which are still alive.
I think the most clear point I wanted to make is the Northeast-enrichment in the data. As expected by the fact that the original colony was located there I still find surprising the few presidents the south and the west has had. The Ohio data was also surprising as I thought it was a smallish state.
The second least obvious thing to see is that presidents are becoming older and older for Republicans (i.e. Trump) while younger and younger for Democrats (Carter, Clinton and Obama). I assume this trend will change if Biden is elected who will become the oldest person to become president in the US at 77.
The third observation is how many presidents ended up dead in office. I find surprising that half of them weren't assassinated but died from natural causes.