Is that book cover a data visualization?
Dear Readers,
Let me tell you about an experiment in data visualization.
By now, if you're reading this, you've seen the cover of Democracy's Data. You probably noticed that some of the letters in the title and in my name are either Blue, Gray, or Red. If you're like some of the first people I shared the cover with, you have wondered if there was any pattern in the coloring of the letters. You might have wondered: does it mean anything?
The short answer is: no. It is just pretty.
One of the people who was disappointed by that answer was Robin Sloan. So he and I put our heads together. What if we could redesign the cover to reveal something important about American democracy?
For a while it seemed like a fool's errand. Then I got a twelve-year-old to help. Talking it out with the Bouk child, who is a lover of books that involve solving codes, I realized a tantalizing fact: "Democracy's Data: The Hidden Stories in the US Census" and "Dan Bouk" together contain 50 letters. Isn't that convenient?
CLICK HERE for the final result. (Some of you have seen the visualization before in an earlier newsletter.) You'll also find an explanation of the visualization paired with some discussion of all the recent fuss in NY over redistricting, culminating in a case for expanding the House of Representatives.
Having completed the visualization, we wanted to make it fun. My editor at MCDxFSG, Sean McDonald, suggested a contest and he offered the prize: whoever broke the code first would get their name printed in every single copy of Democracy's Data. That sounded pretty cool to me!
So: I tweeted out the visualization as a gif, along with the contest premise, and I waited...
And waited some more...
Twitter's algorithm did not seem to like gifs, or it didn't like me.
Then a magical thing happened. A fellow historian of not-actually-boring numbers, John Handel, decided to share the puzzle with his class of ~30 first-year students at the University of Virginia. They were studying empirical methods: how to learn to think with and about evidence. They set out to reverse-engineer our visualization.
Students broke into teams. They took screenshots, slicing the visualization into its component parts. They figured out that the letters likely corresponded with states, and then they started generating hypotheses. They went looking for correlations with other data sets: records of election results, the outcomes of each census, etc. (They were getting very warm!) That class session ended, but the work didn't. These UVA students kept going, sharing leads and ideas well into the evening. (John kept me abreast of it all, much to my delight.)
Finally, after a couple more weeks of work, the class settled on a solution. SPOILER ALERT :-)
I received this photograph in a text message:
Later that day I beamed into their class over zoom and they explained their process. Two elements of the visualization had forced them to revise their early theories. First, they noticed that every time a letter appeared, it was blue. That fact ruled out any electoral interpretation: neither Dems nor Republicans always won when a state first joined the union. Next, they focused on the moment when all of the letters turned gray (or black). Once they figured that part out, they knew they had it. (And they wisely used some research into my past publications to garner clues.)
John had done a wonderful job teaching these students. The entire experience was delightful, and impressive. I tasked them with coming up with a corporate title to use in the book, since we hadn't allocated space for printing ~30 individual names.
A week later, John shared with me the name they had voted to use. He was not thrilled with it. But I deemed it fittingly playful. This is what will appear in the acknowledgements when you open your copy of Democracy's Data in August:
A group of University of Virginia students who elect to be known as "Prof. Handel's Children" cracked the cover code.
I'll close with some other book news. DD got its first review, from Kirkus Reviews one of the big publications that speaks to booksellers and publishing people more generally.
Here are a couple highlights:
"A page-turning examination of why we need to understand the census and its wide-ranging effects."
&
"Bouk shines a bright light on the power of the data to be used as a tool to promote or silence the voices of certain demographics."
It's page-turning, meticulous, a deep-dive, important, and convincing.
I'll take it!
I'm excited to hear what each of you think, once you've seen the book for yourselves. You can e-mail me, of course, or share your reaction online, where it might attract other readers!
Thanks for reading! As always, please feel free to share this newsletter with a friend.