Q1.
In the three examples we considered — the heights, short-stories, and newspaper pages — observe the variability in data when: (a) the mean and median are close to each other (b) the mean and median are comparatively far apart, with mean < median (c) the mean and median are comparatively far apart, with mean > median
Answer
Each of the three examples illustrates exactly one of the three cases.
| Case | Example | Mean | Median | What the data looks like |
|---|---|---|---|---|
| (a) close together | Newspaper pages 10, 16, 16, 18, 20, 22, 26 | 18.29 | 18 | Values spread out evenly, no value far from the rest |
| (b) mean < median | Poovizhi's family 118, 165, 170, 173, 175 | 160.2 | 170 | One value far below the rest — a low outlier |
| (c) mean > median | Short stories 0, 0, 1, 2, 3, 5, 5, 6, 7, 8, 8, 10, 12, 15, 40 | 8.13 | 6 | One value far above the rest — a high outlier |
Why it happens: the median is anchored to the middle position and cannot be dragged. The mean is a balance point, so it slides towards whichever side the extreme value is on. A low value drags it down and it falls below the median; a high value drags it up and it rises above the median. When nothing extreme exists, there is nothing to drag it and the two sit together.
Tip: you can use this backwards. If you are told only that a set of exam marks has mean 42 and median 55, you already know that a few very low marks are pulling the mean down — without seeing a single mark.