Mean, median and mode

In short: Mean, median and mode are three ways to describe a set of numbers with one typical value. Which fits depends on the data.

In more detail: The mean is the sum divided by the count. The median is the middle value of the sorted list. The mode is the most frequent value.

In Depth

• Example: Salaries 2,000, 2,100, 2,200, 2,300, 10,000. Mean 3,720 €, median 2,200 €. The single outlier pulls the mean up, the median stays typical.
• Median with an even count: Mean of the two middle values.
• When mean? For symmetrically distributed data without strong outliers, such as body heights or measurement errors.
• When median? For skewed distributions such as incomes, prices or response times.
• Mode: Also meaningful for categories (most common colour, shoe size).
• Weighted mean: Each value counts with its weight, for example grades by credit points.
• Percentiles: The 95th percentile says below which value 95 % of measurements lie. In IT it describes response times (spread).
• Pitfall: “Average” is ambiguous. Always ask whether mean or median is meant.

See also: correlation and causation