Mean, Median, Mode & Range

The mean uses every reading, so one odd value drags it. Push a single reading far out and the mean chases it while the median barely moves. That sensitivity is why a mean can be misleading when a set contains an outlier, such as one faulty measurement.

The median only cares about position, not size. Put the values in order and take the middle one; with an even count, average the middle two. Because it ignores how extreme the ends are, it is the more honest average for skewed data.

Range measures spread, not centre. Two sets can share the same mean and be completely different: one tightly grouped, one scattered. Quoting an average without a spread hides that, which matters when the readings are tolerances or test results.

• Try this: Drag one reading up to 40. Which average moves most, and which hardly moves at all?
• Try this: Make three readings the same value. What is the mode, and how does it compare with the mean?
• Try this: Arrange the readings so the mean and median are equal. What does the set look like?