A chart of estimates can look more definite than the underlying information supports. Showing a range can help, provided the reader understands what that range represents.
Label the interval and explain the method used to produce it. A range of observed values, a confidence interval, and a prediction interval describe different ideas. Similar-looking marks do not make them interchangeable.
Read the interval alongside the central value and the context of the question. Avoid drawing a strong conclusion from a small visual difference without considering the method and its assumptions. A useful range makes uncertainty easier to discuss, rather than merely adding decoration around a point.
A small working example.
A planning chart can show a range of likely task durations beside its central estimate. The reader sees why an exact-looking schedule may need flexibility.
A note to keep beside it.
A column name can hide an assumption. Check units, missing values, and the counting rule before combining observations that only appear to be alike.
- Explain what the interval represents.
- Keep the method discoverable.
- Consider uncertainty before interpreting a small difference.
Follow a related question
Define the start and end of the event.
Timing an ordinary eventAsk which situations the examples cover.
Examples shape the model’s taskKeep learning
Related background to continue exploring this subject.
NIST: statistical engineering W3C: accessibility, usability, and inclusion
