Use multi-vari charts as a preliminary tool to look at interactions among your data. An interaction occurs when the change in response from one level of a factor to another level depends upon the level of another factor. To test the significance of this interaction, you could further analyze the data using analysis of variance or general linear model techniques.

The chart displays the means at each factor level for every factor. For example, in a two-factor (Factor A and Factor B, each with two levels) multi-vari chart you would see the following:

·    Blue symbols represent the means of the first factor at each level of the second factor. The blue lines connect the means of the first factor within each level of the second factor.

·    Red symbols represent the means of each level of the second factor. The red lines connect them.

For the fertilizer data:

·    Blue symbols represent the means for each Fertilizer level at each level of Variety. The blue line connects the factor level means. Here, the amount of Fertilizer has a greater effect on Variety 1 compared to Variety 2.

-    For Variety 1, low levels of fertilizer (Fertilizer 1) results in a lower yield than high levels of fertilizer (Fertilizer 2).

-    For Variety 2, low and high levels of fertilizer have about the same effect on the yield.

This is evidence of an interaction between Fertilizer and Variety.

·    Red symbols represent the means for each level of Fertilizer. The red line connects the Fertilizer level means. Here, Fertilizer 2 results in a greater average yield than Fertilizer 1.

Thus, the scientists might conclude that there is an interaction between Fertilizer and Variety. The new variety (Variety 2) appears to result in about the same yield regardless of whether a low or high amount of fertilizer is applied.