Show the mean explicitly when readers must judge the mean

Show the mean directly when the chart asks readers to compare group averages instead of making them infer it from multiple raw values. For example, add a visible mean mark or mean value for each group rather than relying on many bar tops, lengths, or areas to imply the average.

Inferring a mean from several marks forces readers to aggregate visually, and the experiments show that they can substitute simpler proxies instead. A direct mean representation turns the task into reading the intended summary rather than reconstructing it from raw marks.

Mechanism: Explicit mean encoding reduces the chance that readers will rely on summed area, total length, or other proxies when answering an average-comparison question.

Evidence: Viewers comparing averages from groups of bars often behaved as if they used summed area rather than the true average, and the paper identifies explicit representation of the mean as a direct design implication when the graph requires mean judgments (Yuan et al., 2019).

Notes: This recommendation is stated as a design implication from the experiments rather than as a separately tested mean-marker condition.

• User Goal: Decide which group has the higher mean.
• Task: Compare average values across multiple observations.
• Data: Quantitative groups with multiple values per group.
• Chart Setting: Displays that currently show raw bars or other raw marks without a direct mean readout.
• Success Criterion: More accurate reading of which group mean is larger.

Break it when: The reader’s task is about individual values rather than a group average. Why: An explicit mean does not address the main question in a single-value comparison.

Sacrifice: You add a derived summary element to the chart. Risk: The chart now carries both raw values and a summary, which changes the display from raw-only to mixed raw-and-summary reading. Mitigation: Use the mean marker when the intended judgment is explicitly about the average.

Mistake: Leaving the mean implicit in a multi-value comparison and expecting readers to average the raw marks accurately. Why it fails: Readers may substitute summed area or length for the intended mean.

Failure Sign: Readers must scan several raw marks and mentally average them to answer the chart’s main question. Quick Check: Ask whether each group’s mean can be read directly from the chart without combining multiple raw values. Stronger Test: Compare the same chart before and after adding an explicit mean representation and see whether the mean judgment becomes easier.

• Add one explicit mean representation for each compared group.
• Make the mean a directly readable part of the chart instead of leaving it to visual aggregation.
• When raw bars remain, use the mean representation to answer the average-comparison question directly.