Statistics for Everyday Life
Mean vs Median
The mean and the median are both called "the average," but they describe the center of a data set in different ways — and on lopsided data they can tell very different stories.
| Aspect | Mean | Median |
|---|---|---|
| What it is | The arithmetic average | The middle value when sorted |
| How to find it | Add all values, divide by how many | Sort the values, take the middle one |
| Effect of outliers | Pulled strongly toward extreme values | Barely moved by extremes |
| Best for | Symmetric data without big outliers | Skewed data, or data with outliers |
| Everyday example | Average income, inflated by a few high earners | Median income, the typical earner |
When to use mean
Use the mean when the data is roughly symmetric, has no extreme outliers, and every value should count — for instance when the total matters.
When to use median
Use the median when the data is skewed or contains outliers, such as incomes, house prices, or response times, where a few big values would distort the mean.
Frequently asked questions
- Is the mean or the median better?
- Neither is universally better — it depends on the data. For symmetric data they’re close and the mean is fine. For skewed data or data with outliers, the median usually describes the typical case more honestly.
- When are the mean and median equal?
- They coincide when the data is perfectly symmetric, like a normal distribution. The more skewed the data, the further apart they drift, with the mean pulled toward the long tail.
- Why is median income reported instead of mean income?
- Because income is right-skewed: a small number of very high earners drag the mean upward, making it look like the typical person earns more than they do. The median stays at the middle of the pack.