The data visualization above was created using Wordle. I decided to use this site because I liked the finished product better than what I created with some other tools. For example, Voyant and DocuBurst both produced visualizations that didn’t appeal to me (Voyant seemed too simple, while DocuBurst was overly complex). When I used Wordle instead, I was impressed with the word size to word frequently relationship, as well as the ‘Randomize’ feature. With the click of a button I could see an entirely new color scheme and arrangement of the words, but the same data being conveyed. I happily clicked away until I found the visualization that I wished to base my analysis on, which is what you see above. I picked a white background with various text colors as opposed to the dark background with light text colors because I found it more clean looking and easier to read.
I really liked the idea of word frequency being proportional to word size on the visualization because of the area of the book I decided to analyze: the chronology. In my head, using Wordle’s capabilities in conjunction with a timeline from 1863-2063 would create an interesting “history book.” Whether or not this actually worked is up for interpretation, but I would argue that it did to an extent. As I discuss in my analysis, many of the largest words have to do with war and death (go/goes are used to represent a death in the chronology). Sam’s narrative starts midway through the American Civil War, and many wars follow during the time period of Only Revolutions. In that sense, my data visualization represents history in the way I envisioned. On the other hand, other words (particularly Derby) which appear frequently do not make as much historical sense to me.
Apart from the historical timeline interests, the main factor while deciding what visualization to use was aesthetics for me. Like I said, I preferred the dark on light color scheme as opposed to the reverse, but I also took into account how easily readable the data was. On some of the graphics, it was difficult to tell what any of the words smaller than “go” and “goes” were, so I continued to randomize the results until I found one where several of the smaller words were legible, as I wished to include as much data in my analysis.
Something interesting I realized during my analysis was how many data points I actually had included in my visualization. Using Voyant, I determined that the chronology included 38,469 data points, which were individual words for my creation. I never could have imagined myself creating such a simple data graphic using tens of thousands of data points – I can only imagine Edward Tufte would be proud of me. In all seriousness, this assignment has provided me the opportunity to delve into some tools and techniques that I may never have otherwise discovered, and hopefully I will be able to incorporate these into my future work.