Guidelines
Suggest edit

Add discrete interval extrema summaries for minimum and range comparisons

For minimum and range comparisons over grouped time intervals, use explicit extrema summaries on position-based time-series charts to improve fidelity and mitigate errors from reading interval lows and spans off an unsummarized line.

  • purpose:refine
  • basis:empirical
  • task:compare
  • time:time-interval
  • data:temporal
  • quality:fidelity
  • lever:encoding
  • channel:position:use

advice

Explicit highs and lows per interval

Add explicit interval highs and lows when readers must compare minima or ranges across time blocks. For example, layer monthly range bars onto a line graph as in a modified stock chart, or show interval whiskers in a box plot, instead of forcing readers to infer lows and spans from the raw line alone.

reason

Why explicit extrema summaries work

Minimum and range judgments require readers to find and compare multiple points inside each interval. Directly encoding the interval high and low reduces that search and gives the reader a clearer monthly comparison unit.

Mechanism: Explicit extrema summaries turn a multi-step visual search into a direct interval comparison, especially for tasks that depend on the local minimum or on the gap between local minimum and maximum.

Evidence: For minima and range tasks, position encodings that explicitly encoded local extrema, such as modified stock charts and box plots, outperformed plain line graphs and other encodings without those summaries (Albers et al., 2014).

context

Use when the reader compares interval lows or spans

  • User Goal: Find the interval with the lowest point or the largest max-minus-min range.
  • Task: Compare extrema-derived summaries across known time blocks.
  • Data: Quantitative time series partitioned into discrete intervals.
  • Chart Setting: A position-based time-series chart where interval summaries can be added.
  • Success Criterion: Higher accuracy on minimum and range judgments.

exceptions

Do not use this as a blanket fix for maxima

Break it when: The primary task is comparing interval maxima with box plots alone. Why: The paper reports that box plots did not outperform the other position encodings for maxima, and it notes possible whisker-related judgment biases.

costs

Costs of adding extrema summaries

Sacrifice: You add more visual structure to the chart. Risk: A layered summary such as a modified stock chart can become visually cluttered. Mitigation: Encode only the extrema summaries needed for the task instead of layering many different summaries at once.

mistakes

Common failure with this lever

Mistake: Leave viewers to recover interval lows and ranges from a plain line graph. Why it fails: Readers must search each interval for two separate points and then compare those derived values across intervals.

check

How to test the choice

Failure Sign: Reviewers scan back and forth across the raw line to estimate interval lows or spans. Quick Check: Show the same data once with explicit interval highs and lows and once without them, then ask which interval has the lowest value or largest range. Stronger Test: Run both questions on the same A/B pair to confirm that the explicit-summary version helps both tasks.

fix

What to change

  • Add one low marker and one high marker per interval.
  • Draw a range bar or whisker summary for each interval.
  • Keep the original line underneath only if raw local detail is still needed.
  • Replace the raw interval segments with a box-plot summary if clutter becomes a problem.

References

Albers, D., Correll, M., & Gleicher, M. (2014). Task-driven evaluation of aggregation in time series visualization. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 551–560. https://doi.org/10.1145/2556288.2557200