Mean
The mean is what most people call the average. Add every value and divide by how many values there are:
For the data 10, 2, 38, 23, 38, 23, 21 the sum is 155 and there are 7 values, so the mean is 155 / 7 ≈ 22.143. The mean uses every data point, which makes it a good summary for roughly symmetric data. Its weakness is sensitivity to extreme values: replace the 38s with 380s and the mean jumps to about 120, even though most values are still in the 20s.
Besides the arithmetic mean, the calculator shows the geometric mean, the nth root of the product of the values. It is the appropriate average for growth rates and ratios, and it is only defined when no value is negative.
Median
The median is the middle value after sorting the data from smallest to largest. Half the values are at or below it and half are at or above it.
- If the count is odd, the median is the single middle value. Sorted, the example data is 2, 10, 21, 23, 23, 38, 38, and the fourth value, 23, is the median.
- If the count is even, there are two middle values and the median is their average. For 1, 4, 6, 9 the median is (4 + 6) / 2 = 5.
Because it depends only on position, the median barely moves when an extreme value is added. That is why home prices, household incomes and wait times are usually reported as medians.
Mode
The mode is the value that appears most often. A data set can have:
- One mode: in 3, 5, 5, 8 the mode is 5.
- Several modes: in the example data both 23 and 38 appear twice, so the data is bimodal.
- No mode: if every value appears exactly once, no value is more common than another.
The mode is the only one of these measures that also makes sense for non-numeric data, such as the most common shoe size sold or the most popular menu item.
Range
The range is the simplest measure of spread: the largest value minus the smallest. For the example, 38 − 2 = 36. It is quick to compute and easy to explain, but it depends on just two values, so a single outlier can make it misleading. For a more robust picture of spread, look at the standard deviation or the interquartile range.
Worked example from start to finish
| Measure | Data: 4, 8, 15, 16, 23, 42, 8 | Result |
|---|---|---|
| Sorted | 4, 8, 8, 15, 16, 23, 42 | |
| Mean | 116 / 7 | 16.571 |
| Median | 4th of 7 sorted values | 15 |
| Mode | 8 appears twice | 8 |
| Range | 42 − 4 | 38 |
Here the mean sits above the median because the 42 pulls it upward, a sign of right skew.
Which one should you report?
Use the mean for symmetric data without outliers, such as repeated lab measurements. Use the median when the data is skewed or has outliers. Use the mode for the most typical choice or category. Report the range together with a measure of center so readers know how spread out the values are. When in doubt, report more than one: if the mean and median are far apart, that gap itself tells readers the data is lopsided.
Mean, median and mode in a skewed data set
Consider the annual salaries at a small company: five staff earn $42,000, $45,000, $45,000, $48,000 and $51,000, and the owner earns $300,000. The mean salary is $88,500, which describes nobody. The median, $46,500, is close to what a typical employee earns, and the mode, $45,000, is the most common salary. This is why reports on incomes and home prices favor the median: one very large value cannot drag it far. When you summarize your own data, check whether the mean and median tell the same story before choosing which one to headline.
Tips for entering data
Separate values with commas, spaces or line breaks, so a column pasted from a spreadsheet works as is. Decimals and negative numbers are accepted. The geometric mean is not defined when any value is negative and equals zero if any value is zero.