Data Visualization: A Mini-Soapbox

Earlier this week the blog Visualising Data posted its pair of pennies on the role of new data visualization techniques in making sense of the WikiLeaks Afghanistan datasets. Visualising Data (British spelling intended!) is dedicated to advancing the principles of good data design in journalism, academia, advertising, and elsewhere. It’s one of my favorite blogs because it falls on my preferred side – and, I think, the preferred side of virtually all social scientists – of what I call the “pretty pictures v. useful pictures” debate. That is to say, the useful side.

Consider its brilliant and instantly narrative visualizations of Afghan war casualties:

All Casualties – Combined
All Casualties - Combined

Deaths Mapped
Deaths Mapped

(all images created by Visualizing Data and powered by Tableau)

In the blink of an eye we learn the Taliban has suffered relatively higher losses, that civilian deaths are a heavy component of the fighting, and that the unruly South is indeed the hotbed of violence. But then, by way of illustration, consider the clunker that gets thrown in midway through:

All Casualties – Heat+ Shape Maps
All Casualties - Heat+ Shape Maps

This graphic falls on what I call the “pretty pictures” side of things. Just looking at it, you know that a significant amount of programming time and mathematical know-how went into its creation. The scale shading, the visual proportioning, the color scheme… As someone who barely knows enough R to scrape by, I am thoroughly impressed by the technical artistry that went into this visualization. But it doesn’t really tell us much. How many casualties are being discussed here? What is the relationship between deaths in one category and deaths in another? The easiest way to find out would be to refer to one of the graphs above it. Those are the “useful” ones. This is a “pretty” one.

Data visualization is a vital component of honest communication that can present an argument and tell a story. But in the hands of an eager show-off, visualization practices can sometimes become exercises in programming pissing contests, where the technical wizardry of the piece is matched only by its rhetorical uselessness. Visualizing Data is one of the best blogs out there because it understands this crucial divide.


Tags: , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: