Mean vs Median vs Mode

Mean, median, and mode are three different ways to summarise "the middle" of a dataset. Picking the wrong one can make your analysis wildly misleading.

Mean (arithmetic average)

$\bar{x} = \frac{1}{n}\sum_{i=1}^n x_i$

The mean uses every data point, which is its strength and its weakness. Strength: minimises squared loss, plays nicely with calculus, underlies regression / variance / Gaussian distributions. Weakness: a single extreme outlier can drag it far from the bulk of the data.

Use the mean when data is roughly symmetric and outliers are rare or already removed.

Median (middle value)

The median is the middle observation after sorting. For income, response time, file size, and other heavy-tailed distributions, the median is far more representative than the mean — Bill Gates walking into a pub raises the mean income but barely moves the median.

Use the median for skewed data, when reporting a "typical" value, or when robustness matters.

Mode (most frequent value)

The mode is the value that occurs most often. Mostly useful for categorical data (favourite colour, browser type) where mean and median don't even apply. For continuous numerical data, the mode often doesn't exist in any meaningful sense — every observation is unique.

Which to use

Scenario	Best measure
Symmetric, outlier-free numerical	Mean
Skewed numerical (income, latency)	Median
Categorical	Mode
Reporting "typical" to general audience	Median
Underlying further calculus / stats	Mean

If you are just starting statistics, internalise this: mean for math, median for stories.

Try them yourself

Paste any dataset into our Mean Median Mode Calculator and see all three at once.