In the FlowingData forums, Ryan asks a really good question about data design:
What simple rules should we all follow when we present data?
I came up with three rules of thumb a while back, but surely there are more. Context, clarity, and real data are clear winners, but what else is there? Those are really broad and can be broken down a few ways – like reducing the number of variables could contribute to clarity. If you have any ideas, please do post your ideas to the forum thread.
Ah yes, I can hear you flipping through your Tufte books.
For me, it’s: tell a story. Don’t leave it to the viewer to guess what you’re trying to explain by showing them a graph or whatever.
If you know what exactly you’re trying to explain, you’ll find the most logical way to show it.
I can’t recall where I first stumbled across these, but when designing visualizations I consider these:
* What information does this allow me to compare?
* How effortlessly can these comparisons be made?
tell a story and comparisons – good ones
Ah thank you Nathan for posting about this, hopefully it will generate a good discussion. My two so far (like I said in the forum) were “Make it stick” (even good information shown in a boring or unappealing way will not resonate with an audience) and “Content and Context” (meaning that numbers and data are great, but without a context, they carry less weight.)
Totally – make sure it tells the story you’re trying to tell. I always start from “what’s my story” and then “how do I tell that clearly, succinctly, and with the level of depth/completeness necessary?”
All too often, loosely related “Data” is tossed up to “support” some claim, though it’s not a direct tie to the story.
To quote Nigel Holmes,
1. What’s the point of the graphic?
2. What does the reader need to know?
I second the comment on context. And that means specifying a lot of things: what’s being measured, when, where, and other points.
Remember the difference between graphics for analysis and graphics for presentation. The graphic you saw the answer in may well not be the best graphic for presenting it to others.
If you can find a copy, the 1948 Census Manual of Tabular Presentation remains one of the best basic texts on the subject. It’s a lot more practical than Tufte.
I second the comment on context. And that means specifying a lot of things: what’s being measured, when, where, and other points.
Remember the difference between graphics for analysis and graphics for presentation. The graphic you saw the answer in may well not be the best graphic for presenting it to others.
If you can find a copy, the 1948 Census Manual of Tabular Presentation remains one of the best basic texts on the subject. It’s a lot more practical than Tufte.
The “Manual of Tabular Presentation” is available online here: http://www.archive.org/details/BureauOfTheCensusManualOfTabularPresentation, although unfortunately in the awkward djvu format.
I have made a pdf version with searchable text from the tiff images, but I’m still fighting with acrobat to make it a decent size (it’s currently nearly 100 meg). Please email me if you’d like a copy.