Measures of central tendency give a single value that represents the centre of the data; the three main ones are the mean, median and mode.
1. Arithmetic mean \(\bar x = \dfrac{\sum x}{n}\) (for grouped data \(\bar x = \dfrac{\sum f x}{\sum f}\)).
Uses all values and is best for further calculations, but is affected by extreme values. Example: salaries 20, 22, 25, 23 and 200 thousand have mean 58 thousand, which describes nobody well.
2. Median: the middle value after arranging data in order. For odd \(n\) it is the \(\left(\frac{n+1}{2}\right)\)th value; for even \(n\), the average of the two middle values. Not affected by extreme values, so it is preferred for skewed data such as incomes and house prices.
3. Mode: the most frequent value. Can be used for categorical data (most popular shoe size or brand). A data set can have no mode or more than one mode.
Worked example. Data: 2, 3, 3, 5, 7, 10 (\(n = 6\)).
- Mean \(= 30/6 = 5\)
- Median \(= (3 + 5)/2 = 4\)
- Mode \(= 3\) (occurs twice)
Here mean > median > mode, which indicates a positively (right) skewed distribution.
Other averages. Geometric mean \(= \sqrt[n]{x_1x_2\cdots x_n}\) (growth rates, e.g. average interest rate), harmonic mean \(= n / \sum (1/x)\) (average speeds, rates). For any set of positive values, \(AM \ge GM \ge HM\).
Empirical relation for moderately skewed data: Mode \(\approx 3\,\)Median \(- 2\,\)Mean.