Tuesday, August 23, 2011

Find: Statistical Graphics and Information Visualization

Statistical graphics vs info viz

Statistical Graphics and Information Visualization

Spiral graph

The two differ in who uses them, how they are used, and who consumes them. They have the same goal. It's to better understand data. You'd think that common bond would draw statisticians and information visualization researchers together for ample collaboration, but that isn't the case. You see, each group doesn't quite understand what the other is doing, and that's where intermingling gets tricky.

In the most recent Statistical Computing and Graphics newsletter [pdf], two short articles — one from a computer science point of view and the other from statistics — contrast statistical graphics and information visualization, respectively.

In the former, Robert Kosara argues the usefulness of InfoVis, namely it's not just pretty pictures and static graphics. InfoVis promotes exploration:

And yet, visualization is much, much more than what it appears to be at first glance. The real power of visualization goes beyond visual representation and basic perception. Real visualization means interaction, analysis, and a human in the loop who gains insight. Real visualization is a dynamic process, not a static image. Real visualization does not puzzle, it informs.

In the latter, Andrew Gelman and Antony Unwin argue the benefits of traditional statistical graphics:

In statistical graphics we aim for transparency, to display the data points (or derived quantities such as parameter estimates and standard errors) as directly as possible without decoration or embellishment. As indicated by our remarks above, we tend to think of a graph as an improved version of a table. The good thing about this approach is it keeps us close to the data.

Wait. Those sound kind of similar. Both articles, written independently of the other, discuss different approaches to visualizing data, but they have similar sentiments.

Oh, but the difference. There has to be a difference.

Kosara uses a spiral example (above) as interaction with data. It shows periodicity.

You can try an interactive version here. I'm still on the fence on the spiral's usefulness, but it has its merits.

Gelman, despite always starting and ending his critiques with a desire to collaborate and learn,

Sent from my iPhone