How Significant is a Boxplot Outlier?

Abstract

It is common to consider Tukey's schematic (“full”) boxplot as an informal test for the existence of outliers. While the procedure is useful, it should be used with caution, as at least 30% of samples from a normally-distributed population of any size will be flagged as containing an outlier, while for small samples (N<10) even extreme outliers indicate little. This fact is most easily seen using a simulation, which ideally students should perform for themselves. The majority of students who learn about boxplots are not familiar with the tools (such as R) that upper-level students might use for such a simulation. This article shows how, with appropriate guidance, a class can use a spreadsheet such as Excel to explore this issue.

• Boxplot
• Outlier
• Significance
• Spreadsheet
• Simulation