Plotting many variables at once

Understanding Data Visualization

Richie Cotton

Data Evangelist

When should you use a pair plot?

• You have up to ten variables (either continuous, categorical, or a mix).
• You want to see the distribution for each variable.
• You want to see the relationship between each pair of variables.

When should you use a correlation heatmap?

• You have lots of continuous variables.
• You want to a simple overview of how each pair of variables is related.

¹ Rossi, Allenby, and McCulloch (2005). Bayesian Statistics & Marketing

The United Nations dataset again

When should you use a parallel coordinates plot?

• You have lots of continuous variables.
• You want to find patterns across these variables, or
• You want to visualize clusters of observations.

A parallel coordinates plot

Let's practice!

Understanding Data Visualization