Avoiding Pitfalls of the Pie Chart in 21st-Century Data Visualization
A recent article in SEEDMAGAZINE.COM laments that many people misuse and rely too heavily on static data visualizations, such as the pie chart. "The pie chart is intended to display proportions of a whole within a single, small data set, but overzealous Excel users dump in large data sets or stack multiple pies. The resulting complex defeats the purpose of using a picture: simplification," writes Veronique Greenwood in "Getting Past the Pie Chart."
The SEEDMAGAZINE.COM article quotes big names in data viz, such as Colin Ware (Director of the Data Visualization Research Lab at the University of New Hampshire and author of Visual Design for Thinking) and Bill Cleveland (statistican at Bell Labs and Professor at Purdue University and author of Visualizing Data) in its useful discussion of how to improve new forms of data visualization. Although the article doesn't include any visuals, the author makes some good points:
"One way to solve the problem of overly complicated diagrams is to introduce interactivity."
According to Cleveland, "an effective way of detecting patterns in massive data sets is to make a simple chart for each subset and view the hundreds of charts in quick succession."
"Numerical tools — statistical tests of variance and significance — are just as important in assessing trends" (also attributed to Cleveland).
These ideas are in line with the way JMP approaches data visualization. "JMP emphasizes interactivity over perfectly styled static graphs and the use of statistics alongside data visualization," says Xan Gregg, a JMP developer and data visualization expert.