Description
Patterns can be many things – a spike on a trend of products at a certain time, or clustering of objects in a bubble chart showing a pattern of commonality. A primary goal of patterns is to discover cause and effect.
But when displaying data with the purpose of showing patterns, you want to make sure that your users are actually seeing those patterns. In order to do so, it’s important to know how to emphasize and call attention to them.
One way to do this is by showing a view that allows one to see multiple sets of data at once. When looking at all of them together, a pattern may be seen. Examples of this include small multiples and whisker plots. Both have examples shown below.
Another way would be to highlight certain information that reveals a pattern with the use of contrast, shape, movement and/or size.
Take the wind map from below for example. Through the use of contrast and movement, wind all around the country is shown in patterns. On the other hand, you could also take multiple wind maps and place them side by side in a small multiples format and show patterns of wind over time.
Despite your approach, make sure the pattern stands out and that the use of visualization and layout complement your desire to show patterns.
Examples
Small multiples – Zach from Juice discusses the comparison tradeoffs in this chart type which simplifies pattern recognition. Here is another example of these as trends.
Wind Map – This beautiful visualization by the good folks at hint.fm shows patterns by plotting all wind data simultaneously as it happens. The culminating shape and speed of the wind itself is comparable from region to region.
Win-loss whisker plot – Here is another type of small multiple that allows one to quickly scan patterns between two options. In this case it is wins(blue) or losses(red) or ties(grey) of the 2011 MLS season.
Related Principles
All on screen
Track changes
Emphasize the interesting