David McCandless’ TED talk is very specifically on data visualization and can be found here. He’s not quite as engaging as Gosling but his ideas are compelling as are his visualizations. Both presenters talk about the importance of representing data in context, for example by showing rates. To make this point, McCandless displays a graphic comparing the total US military budget with that of other countries. Not surprisingly, the U.S. military budget dwarfs that of much of the rest of the world. McCandless’ point however is that it would make more sense to look at military spending as a percent of GDP. Figured this way, US spending looks eminently reasonable, compared say to that of Burundi or Eritrea. But is it? In the US, military spending is about 4% of GDP, in Finland it’s about 1.5%, which country is better off? I don’t have the figures but perhaps a better metric would be the value of the property the defense budget is predicting. That’s how an insurance company would figure it. Or maybe a measure of personal well being (they are starting to do this in the UK as a mater of public policy) or sense of security? That would be interesting… well being as a function of military spending! The point here though is to show that the denominator of the fraction you use has to be chosen with care.
This idea is echoed in a recent article in the New York Times about data visualization: “The purpose of visualization,” says Ben Shneiderman, founding director of the Human-Computer Interaction Laboratory at the University of Maryland, “is insight, not pictures.”
So there is more to data visualization than just pretty pictures. In this post, and the last one, we’ve touched on the story the visualization is designed to tell. But there is also a psychology of visual perception to consider. The use of lines and colors and areas plays a part in how attractive a visualization appears, but also, it turns out, in our ability to apprehend its message. For this we have to look elsewhere. [Read more…]