One of the key pillars of my approach to visualizing data is finding ways to break up a dense chart into smaller chunks. Known most commonly as small multiples—but also as grid charts, panel charts, facets, or trellis charts—these visualizations are smaller charts that use the same scale, axes, and scope but spread the data across multiple visuals. In other words, instead of having all the data on one graph, the creator designs multiple, smaller versions breaking out the basic data.
But implementing the small multiples approach doesn’t mean you need to use a single graph for every series in your data. A simple example comes to mind: performance of sports teams over the course of a season.
Take Major League Baseball, for example. From the beginning of the season in April to the end in October, each team plays 162 games. We can chart their success using the number of games each team plays with more wins than losses. However, plotting all the teams on the same graph would create a mess of lines (all graphs made by Greg Stoll, and you can visit the interactive versions and inspect the code on his site):
Even in the interactive version, this graph is hard to read (except for the historically bad Chicago White Sox at the bottom of the graph).
If we wanted to make the data more readable by using a small multiples approach, we need not create a single graph for each of the 30 teams. While doing so would make it easier to see each team individually, it would take up a lot of room and make it difficult for our readers to explore the data. Furthermore, teams’ performance relative to each other is important to determine playoff seeds and division winners. Thus, a more useful, readable, and meaningful organization would be to create a set of small multiples for each division, such as this one for the National League East division (again from Greg Stoll’s site).
This grouped small multiples approach can be used in different scenarios. When working with your data, consider whether you can group your data into different categories to draw out key takeaways or conclusions. For example, if you’re making a map showing spending changes over time, you could create line charts for states that are increasing, decreasing, or unchanged; four separate maps for areas north, south, east, or west; or three bar charts for urban, suburban, or rural areas.
Small multiples can enable us to show lots of data in clearer and more digestible ways, but it’s important to remember that you don’t need to break up your dense data into individual charts. Groupings might not only be clearer, but also more useful for your readers.
If you liked this post, consider signing up for the PolicyViz Substack newsletter. You get previews of new posts, special features and deals, and interesting books, blogs, and things to watch.