« « How to “Group By” in Excel

Ben Fry on visualizations and Processing » »

Keep it subtle, stupid: Differentiating data values in visualizations

Posted by on Jan 25, 2010 in Blog, class, tufte | No Comments

I’m not anything resembling a visual genius, but I like to believe I have enough of an eye that I can tell what works and what doesn’t.  The reason Tufte’s books are so helpful is that they tell me why something works, and give me rules so I know what I can and can’t do.  That’s why I have great appreciation for the Tufte chapter of the week: “The Smallest Effective Difference” in my ongoing exploration of “Visual Explanations.”  (I’m also reading his other books on the side, and more on that in the coming week.)

The chapter at hand was surprisingly short — six pages — but for the subject matter, it worked.  The “smallest effective distance” refers to how you differentiate between different levels of data in intensity-based charts, or different variables.  The point is, either way, if you make things too distinct, it can be jarring to the eye, so it’s important to keep colors within the same general theme.  At the same time, pick colors that are too close together, and you risk losing the person looking at your information — they should get it at a glance.

It’s true – the most jarringly rainbow visualizations I can think of are the USA Today weather map and this other visualization. They both use rainbow color schemes, but the intensity of the colors are all equally bright, and the values of similar color are all in the same area of the visualization.  Tufte demonstrates that the problem is when a patch of bright red is surrounded by bright yellow, bright pink, and bright green — that’s just jarring.

The latter part I understood before, but hadn’t really thought about the former.  It’s a valuable point, one we use in writing as well.  Don’t jump all over the place, stay on topic, keep a focus to your piece.  This is just a question of keeping a visual focus, and making distinctions easy for the eye to grasp. It’s a simple but powerful lesson.

« « How to “Group By” in Excel

Ben Fry on visualizations and Processing » »