Outlier Detection (the 1.5xIQR rule) with Tableau

A common rule to identify outliers is the 1.5*IQR rule, meaning any data points that are more than 1.5*IQR above the Q3 (the third quartile) or below Q1 (the first quartile).

In this post, I am going to show you how to implement this rule in Tableau to identify outliers, an important step in data exploration.

Background made with Canva — my one-and-only tool for any design needs.

Compute Q1, Q3, and IQR

Use window_percentile function to compute the third quartile and first quartile for the entire table. Use Q3-Q1 to get IQR.

Compute Q1–1.5IQR as the lower bound

Compute Q3+1.5IQR as the upper bound

The Outlier Flag

Specify what constitutes an outlier for the target metric.

Indicate Outliers and Bounds in Chart

• Use reference band to the lower bound to Q1–1.5IQR and upper bound to be Q3+1.5IQR
• Use Outlier flag for color and shape

Club members can download the dashboard template and sample data here for free.

Originally published at https://www.goodmarketing.club on May 22, 2021.