The mean is the most familiar average, but it is not always the most honest one. When a few values are much larger or smaller than the rest, the mean gets pulled toward them and stops describing a typical value. In those cases the median usually tells you more.
You can compare both for any list of numbers with our average calculator.
Try the Average Calculator. Mean, median, mode and range of a list
Open the calculatorHow outliers affect the mean
The mean uses the size of every value. One very large number raises it a lot. The median only depends on which value sits in the middle, so it barely moves.
Example: house prices on one street
Eight homes sold for (in thousands): 280, 295, 310, 315, 320, 335, 350, 1,900
- Mean: 4,105 ÷ 8 = 513
- Median: (315 + 320) ÷ 2 = 317.5
Seven of the eight homes sold for under $351,000, yet the mean says $513,000. The median gives a far better idea of what a typical home on the street costs.
Skewed data
Data is skewed when it trails off much further in one direction than the other.
- Right-skewed data has a long tail of high values. Incomes, house prices, hospital bills and social media followers are classic examples. The mean is higher than the median.
- Left-skewed data has a long tail of low values, such as scores on an easy test where most people did well and a few did badly. The mean is lower than the median.
When the mean and median are far apart, that alone tells you the data is skewed.
Common places the median is used
- Household income. Government statistics agencies usually report median income because a small number of very high earners would inflate the mean.
- House prices. Property reports usually quote median sale prices.
- Response or waiting times. A few very long waits can distort the mean.
- Ages in a group, when there are a few very old or very young members.
When the mean is better
The median is not always the right choice:
- Balanced data. Heights, test scores in a typical class, or measurement errors are often fairly symmetrical. The mean and median are close, and the mean uses all the information.
- When totals matter. If you need to know total cost or total output, the mean connects directly to the total: mean × count = sum. The median does not.
- Further calculations. Standard deviation, many statistical tests and grade averages are all built on the mean.
Use both
Reporting both numbers is often the clearest option. "The median salary is $48,000 and the mean is $61,000" tells the reader that most people earn around $48,000, but a minority earn a lot more.
A quick test
Ask yourself: if I removed the single largest value, would my "average" change a lot? If yes, the median is probably the safer summary.
For the basics of all three averages, see mean, median and mode explained.
Was this page helpful?
Thanks for telling us.
What was missing or wrong? Email us and we will fix it.