A matrix plot is a graph that you can use to assess the relationship among several pairs of variables at the same time. A matrix plot is a set of individual scatterplots. There are two types: matrix of plots and each Y versus each X.

A matrix of plots displays a plot for each possible pairing of variables. A matrix of plots is useful when you have many variables and you would like to examine relationships between all pairs of variables.

A business analyst wants to look for relationships between several different business metrics for manufacturing companies. The analyst collects data on the number of clients, rate of return, and years they have been in business for 22 successful small-to-medium sized manufacturing companies.

In this graph, plots 1 and 4 show that the number of clients and the number of years the company has been in business seem to increase together. The other plots show a possible weaker positive relationship between the variables. The following plots are flipped images:

- 1 (y=Clients, x=Years) and 4 (y=Years, x=Clients)
- 2 (y=Clients, x=Rate of Return) and 5 (y=Rate of Return, x=Clients)
- 3 (y=Years, x=Rate of Return) and 6 (y=Rate of Return, x=Years)

An each Y versus each X plot displays a plot for each possible Y-X combination, when you specify the y variables and x variables. This type of matrix is effective when you are only interested in relationships between responses and predictors, which you enter separately. These plots are also known as draftsman plots or casement displays.

A human resource manager wants to evaluate an employee training program. The manager creates an each Y versus each X plot to examine the effect of work experience, instruction time, and study time (x variables) on test scores and retention six months later (y variables).

This graph shows the following:

- Test scores and retention both increase as work experience and instruction time increase (1, 2, 4, 5).
- The relationship between study time and retention or test scores seems to be weaker (3, 6).