[This is a guest post by Jon Schwabish* and Severino Ribecca**, about the informational poster The Graphic Continuum]
How many different graph types exist? How do they relate to one another? Can you use the same graphic type for different types of data? These are the questions that we tried to tackle in our recent project, The Graphic Continuum.
Documenting the many chart types is something we were both working on independently for the past couple of years. Severino was building his Data Visualisation Catalogue, an online reference tool of data visualizations. At the same time, I was teaching data visualization to different audiences and was thinking about how to best show my students different graphic types and how they relate to one another.
Not too long after Severino’s Data Visualisation Catalogue launched, we got together to work on a static version of a library of different graphic types. We wanted to take this idea of a network of graphic types and convert it to something tangible that people could reference while they worked with their data.
Initially, we visualized how each of the chart types relate to each other: How, for example, you could show comparisons using a scatterplot or a parallel coordinates plot, or how a stacked column chart can help you compare across and within categories. But our first drafts were much too busy and the connections between chart types were unclear. We added, and then removed, the visual properties of each chart such as whether they displayed values via position on an axis, by area, or by angle.
In the end, we decided to group charts based on their main functions and what they are primarily trying to communicate: Comparing Categories, Distribution, Geospatial, Part-to-Whole, Relationship, and Time. Of course, this is a simplification of the true space of data visualization types: a slope chart connects categories across vertical axes instead of showing all points as in a line chart. Our first Graphic Continuum project–a 24″ x 36″ poster–explicitly shows some of these connections.
A few weeks after the poster version was released, and then having won a bronze medal at this year’s Information is Beautiful Awards, we responded to requests for a smaller version. The more compact (8.5″ x 11″) version of The Graphic Continuum fits fit nicely onto your desk to use as an easy reference in your data visualization projects.
Although the desktop version of The Graphic Continuum doesn’t show the connections between categories, we view it more as a resource and less as data art than the poster version. We hope you will view them both as beautiful presentations of the space of graphic types, but also an invaluable resource as you work to visualize your own data.
*Jon Schwabish is a Senior Research Associate in The Urban Institute’s Income and Benefits Policy Center. I am also a member of the Institute’s Communication team where I specialize in data visualization and presentation design. Prior to my time at Urban, I spent the previous 9 years at the Congressional Budget Office conducting research in such areas as earnings and income inequality, immigration, disability insurance, retirement security, data measurement, the Supplemental Nutrition Assistance Program (SNAP), and other aspects of public policy. You can follow Jon’s updates on Twitter (@jschwabish), and keep up with his posts on PolicyViz, and audio episodes of the Rad Presenter, a podcast he hosts alongside Stephanie Evergreen. We also interviewed Jon a couple of months ago, and don’t forget to visit HelpMeViz
**Severino Ribecca is a graphic and information designer interested in data visualisation. Currently he’s building an online library of information visualisation methods called The Data Visualisation Catalogue. You can follow the project’s updates at on Twitter (@dataviz_catalog).