In the past few weeks, we’ve been running a series on the tweaks that designers do to charts to make them fit the aesthetics of the graphics they are creating. So far, we showed you some of the ways designers get creative with pie charts, bar charts, line charts, and area based encodings. Maps are not immune to designer’s creative experiments, and they offer some of the most widely varying techniques for adjusting aesthetics. Some of these techniques are just stylizations, and they can result in beautiful maps, but sometimes they severely damage the communication of the map data. Let’s take a look at five of the ways designers tweak maps and dissect if and how they are harming the communication of the data.
- Choropleth Maps are in shaky territory. We are so accustomed to seeing them that we know what to expect from them, but they have serious issues with disproportionate area representation. The balance between familiar and ideal is hard to measure here, but it’s probably safe to use them.
- Proportional Symbol Maps are a great technique. They solve the problem of some areas being disproportionately large, while still providing a geographical reference. Combined with a bit of collision detection and force direction, they don’t even have occlusion problems. Don’t hesitate to use this technique.
- 3D Symbol Maps, unlike their 2D counterpart, are often a bad technique. They can have issues with occlusion and problems with perspective distortion. A good question to ask before making one of these is: “If one of these symbols was a single chart, would it be a good idea?”
- 3D Height Maps are not a good technique. They look extremely dramatic, but they have severe problems with occlusion and disproportionate areas. Never do this to represent data.
- Transit Maps can be great for showing transit type stuff, but their format is often misused. They are great for transit lines because they keep relative positions of the stations, but the data for the paths in-between is simplified. The lines still mean there is a connection between the points, though. And the style is often copied for no apparent reason. Randomly adding lines to a map of points doesn’t accomplish anything. If you use these, think about whether the lines actually show anything meaningful.
There are many other types of maps, and they can be a great way to show data. Maps are one of the most common and useful forms of data visualization. They can be both beautiful and powerful, but they need to be used carefully in order to communicate effectively. (Map regions from wikimedia commons.) Drew Skau is Visualization Architect at Visual.ly and a PhD Computer Science Visualization student at UNCC with an undergraduate degree in Architecture. You can follow him on twitter @SeeingStructure