To determine if there is a measurable difference in performance and perception of reading 2D and 3D charts, we conducted an online study. We tested 2D and 3D versions of bar, column, donut, and pie charts. Participants were asked to use either a set of 2D or 3D charts to answer a series of questions. We automatically recorded the speed and accuracy of their responses. After completing those tasks, participants were asked to rate the charts in terms of overall ease/difficulty of reading, ease of learning, effectiveness, confidence, visually appeal, familiarity with all of the chart types, and the overall appearance of the charts (1 = childish, 5 = professional). For the final question, participants were shown a version of each chart type they saw earlier in the study. They were asked to pick the chart type that they liked the most. To see the experience first-hand, you can participate in the study here: http://www.surveygizmo.com/s3/2767833/UXPA-2016-Graphs
Our findings showed strong performance differences in favor of 2D charts. However, the 2D charts only experienced a slight edge over 3D charts when it came to the ratings. This is likely because participants were not making a direct comparison between 2D and 3D charts. Lastly, we saw a preference for the column chart type over the others when participants selected the one they liked the most. Overall, these findings help support the use of 2D charts to make your data easier to interpret.