Published Jul 23, 2021

SDS 490: Say No to Pie Charts — with Jon Krohn

Jon Krohn critiques the popular use of pie and donut charts for data visualization, emphasizing their aesthetic appeal but functional inefficiency, and advocates for bar charts as a superior alternative for clear and effective data comparison.
Episode Highlights
Super Data Science: ML & AI Podcast with Jon Krohn logo

Episode Highlights

  • Criticism

    critiques the use of pie charts, emphasizing their aesthetic appeal but functional shortcomings. He points out that pie charts and their cousin, the donut chart, often fail to clearly convey data due to the difficulty in distinguishing slice sizes, especially when they are similar or distant from each other 1. This visual ambiguity undermines the primary goal of data visualization, which is to present information clearly.

    The problem with circular representations of data is that it is difficult for our eye to distinguish which slices of the circle are larger than others.

    ---

    Jon suggests that if clarity is the goal, pie charts are rarely the best choice 1.

       

    Alternatives

    For more effective data visualization, Jon advocates for bar charts over pie charts. He explains that bar charts allow for easy comparison of category heights, making them a superior choice for conveying relative magnitudes at a glance 1. This clarity is crucial when the objective is to make data easily interpretable.

    In the vast majority of circumstances, a bar chart, which makes it extremely easy at a glance to compare the relative height of categories, should be your go to plot type.

    ---

    Jon also hints at rare cases where circular representations might be useful, directing listeners to further resources for those interested 1.

Related Episodes