"The average" is not one thing. There are three common measures of central tendency — mean, median, and mode — and they can tell very different stories about the same data. Knowing which to use is one of the most practical skills in statistics.

The three measures

Worked example. Salaries (in thousands): 30, 32, 35, 38, 40, 250.
Mean = 70.8 · Median = 36.5 · Mode = none.
The mean (70.8) suggests a typical salary far above what anyone except one person earns. The median (36.5) describes the group far more honestly. That one big value — an outlier — is the whole story.

The key difference: sensitivity to outliers

The mean uses every value, so a single extreme number drags it up or down. The median only cares about position, so it barely moves. This is why:

SituationBest measureWhy
Symmetric data, no outliersMeanUses all information, most efficient
Skewed data or outliers (income, house prices)MedianNot distorted by extremes
Categorical data (favourite colour, size)ModeOnly measure that works on non-numbers

How skew shifts the mean

In a perfectly symmetric distribution, mean = median = mode. When data is right-skewed (a long tail of high values, like income), the mean is pulled above the median. When left-skewed, the mean falls below the median. Comparing the mean and median is itself a quick check for skew — if they're far apart, your data isn't symmetric.

Which should you report?

Often the honest answer is to report more than one, plus a measure of spread like the standard deviation.

Calculate all three at once

Descriptive Statistics Calculator →Paste your data to get mean, median, mode, standard deviation, quartiles, skewness, and more.

Related reading: Confidence intervals explained · Effect size explained · Choosing a statistical test.