Guidelines
Suggest edit

Use a bar chart instead of a pie chart for part comparisons

For part-to-whole comparison, use a common-scale bar chart on quantitative shares to improve judgment fidelity and mitigate angle-based comparison errors for readers making visual estimates.

  • purpose:select
  • basis:empirical
  • chart:bar:use
  • chart:pie-donut:avoid
  • quality:fidelity
  • lever:chart-family
  • operator:part-whole

advice

Common-scale bars

Replace the pie with bars on a common scale when the task is to compare part sizes. For example, show each part as a bar rather than a wedge, and put the bar scale in percentages or another simple fraction of the whole.

reason

Why common-scale bars work

Bars let readers compare part sizes by position along a shared axis. Pie slices force angle and area judgments, so ordering parts and estimating ratios is less reliable.

Mechanism: Position on a common scale supports more accurate quantitative judgments than angle. That makes it easier to see both the order of parts and how large one part is relative to another.

Evidence: The paper’s position-angle experiment found position judgments on bars substantially more accurate than angle judgments on pie charts, and its redesign example shows the same part-whole data becoming easier to order once wedges are replaced by marks on a common scale (Cleveland & McGill, 1984).

context

Use when comparing parts of a whole

  • User Goal: Compare which parts are larger and by how much.
  • Task: Make visual ratio judgments or rank the parts of one whole.
  • Data: Quantitative parts that sum to a whole.
  • Chart Setting: You are choosing between a pie chart and a bar-based display.
  • Audience: Readers making quick visual estimates.
  • Success Criterion: More accurate part comparisons and clearer part ordering.

exceptions

Do not use when the chart is not for quantitative part comparison

Break it when: The display is not meant to support visual comparison of quantitative parts within a whole. Why: The source supports this substitution specifically for judging relative part magnitudes.

costs

Tradeoffs of replacing the pie

Sacrifice: The display no longer uses a circular whole. Risk: If the bar scale is not expressed as a fraction of the whole, the whole can feel less explicit. Mitigation: Use a 0-100 scale, or another simple fraction scale such as 0-25 or 0-50, so each bar still reads as a share of the whole.

mistakes

Common failure with part comparisons

Mistake: Keep wedges because the chart already “shows parts of a whole.” Why it fails: The comparison still depends on angle and area judgments, so part ordering and percent-of-larger estimates stay hard.

check

Check the comparison directly

Failure Sign: Readers hesitate when asked which part is larger or what percent one part is of another. Quick Check: Render the same values as a pie and as bars, then ask for the part order or a percent-of-largest estimate. Stronger Test: Keep the version that yields quicker and more accurate visual estimates; for this contrast, that is the bar version.

fix

Fix the display

  • Replace each wedge with one bar on a shared baseline.
  • Express the bar scale in percent or another simple fraction of the whole.
  • Remove the pie when the chart must support ordering or ratio judgments between parts.

References

Cleveland, W. S., & McGill, R. (1984). Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods. Journal of the American Statistical Association, 79(387), 531–554. https://doi.org/10.1080/01621459.1984.10478080