I surely needed to be doing something else today, but one of my deepest interests, data visualization, is such a seductive distraction for me — and ever since I saw Hans Rosling’s dazzling TEDTalk on “the best stats you’ve ever seen,” I have been intrigued by graphs that move.
It seems, now, that Google has added the same sort of motion graph widget to Google Docs Spreadsheets, and I started playing around with things that I could demo in an upcoming presentation. I didn’t find anything that was perfect for the presentation, but pulling the “Internet Usage Statistics” up out of my Delicious account gave me an idea.
Unlike most spreadsheet layouts for graphing, the dates are stacked as shown here.
I went to The Way Back Machine and found archived versions of the stats page going back to 2002 — and the 2000 data is available in all of the graphs, as a comparison. Then I began to re-type the data into a Google Docs Spreadsheet. I could have come up with a funky way to import it in, but there was so little data, that I just re-typed. I found a number of irregularities in the data that I could only reconcile mathematically and not historically — but I was doing this for the fun of it anyway.
The layout of the data in the spreadsheet is critical. Counter intuitive is the fact that you have to group the item (region of the world) and the time (2000-2009) on a single row. This way 2000 repeats itself for each region and then 2002 – 2009. To the right on each row is the number of Internet users per region per year and its penetration (percent of the the total population of the region at that time).
I have the chart configured to start with the population along the horizontal axis and the totol number of uses along the vertical. But you can change these easily as well as set individual regions to trace their changes.