It’s important to know the difference between correlation and causation when using charts. Duh. Below is a good example of why.
I like it. It never occurred to me to use colored lines to differentiate rising and falling values in this type of chart. I would suggest sorting by the change instead of the most recent observation, but I suppose it would depend on the point you were trying to make.
While we’re talking about logos… do you agree with these color categories? Some of these brands are quite old. What came first, the color or the emotion?
Some fascinating stuff in here. (And yes, I know these aren’t charts, but I like to also post on interesting graphic design issues.) (via)
I have no idea why I never noticed the 31 before. it’s pretty obvious.
Want more? Google “hidden logos”.
How does your brain process visual information and relationships? Read this article to begin to understand it.
Here’s a related article on the topic by Stephen Few that you might find interesting.
Great video describing some of the technologies that go into making the live coverage meaningful. (via TechCrunch)
All by the same company that brings you those first down lines, strike zones, and nascar labels.
I usually don’t post this stuff, but it’s Friday afternoon, and I like them. Mike Baboon has a great eye for distilling form and creating color palettes.
Lots of books try to convince us that different colors are associated with different emotions and messages, but John Nelson decided to put google image search results to the test. Check out the comments for some good discussion of the strengths and weaknesses of the technique.
There really aren’t many differences between the two current proposals, which were designed to incorporate the new Silver Line to Dulles airport. As mentioned in the WashPost comments, the only major flaw I see is using the same color for the background beltway and DC boundary as for the Silver Line.
To be honest, I still prefer a lot of the options that Cameron Booth proposed in his 2010 reworking:
Every once in a while you discover something that is obviously useful, but for some reason, you never even considered it before. That happened to me today in the form of graph digitizing software. Basically, it lets you create a chart in reverse: it starts with an image of an existing chart, and extracts the data. Why is this useful? Well, sometimes you find a great chart and want to use it for a project, but you don’t have access to the original dataset (it’s lost in time, etc). I’ve gotten around this in the past by tracing images in vector software – but obviously, extracting the data provides more precision and flexibility.
There are several digitizing programs out there, but I like Graphclick because it works with multiple chart types and has a simple interface (though it only works on Macs). Un-Scan-It is a nice one that works on PCs and Macs (but is a bit expensive).
An addictive collection of beautiful charts, graphs, maps, and interactive data visualization toys -- on topics from around the world.