Definition

A graphical tool used to display the centering, spread and shape of a continuous dataset. It essentially gives a pictorial representation of the five point summary descriptive statistics: the minimum value, lower quartile, median (sometimes includes the mean), upper quartile and maximum value. Also known as 'Box and Whiskers' plot because the lines extending from either end of the box to the maximum and minimum are called 'whiskers'.

Examples

Boxplot Example:

Application

In some cases, such as when there may be outliers in the data, a slight modification of the boxplot is used: here, instead of extending to the maximum and minimum, the whiskers of the boxplot extend to the point closest to and within +/-1.5 times the inter-quartile range (IQR = Q3 - Q1). Points falling beyond the whiskers are then flagged as suspected outliers. Points lying outside the +/-3*IQR region are flagged as extreme outliers.

Advantages:
1. Gives a quick graphical view of the data.
2. Side-by-side boxplots, in which the distributions of multiple categories/groups are displayed in parallel, allow for comparisons among the groups.

External Links

Boxplots from NetMBA: - http://www.netmba.com/statistics/plot/box/