Training

You can use the Chart Builder feature that is found under "Graphs" in IBM SPSS Statistics to graph interactions between variables. For example, perhaps you’d like to depict an interaction including three variables: race, sex, and the number of television hours viewed each day.

Begin by selecting a line chart within Chart Builder. Next, choose the Multiple Line Chart option by dragging it onto the canvas. Since you’re interested in whether television viewing varies by race and sex, place race on the x-axis, television hours on the y-axis, and sex into the "set color" box.

Here’s what you’ll see in the Chart Builder window:

To interpret the output, look at the relationship between the lines in the graph. If the lines remain parallel, there is no interaction present. If the lines cross, there is evidence of a strong interaction. Finally, if the lines diverge and do not cross, this suggests a weak interaction.

After graphing the example above, you’ll notice that the lines are parallel, so no interaction exists:

This suggests that males and females watch the same amount of television each day across all racial groups.

The line graph is easy to use and illustrates interaction effects well. Be sure to take advantage of this graph when presenting all of your interaction effects!

*IBM SPSS Statistics and IBM SPSS Modeler, formerly called SPSS Statistics and Clementine®, are part of SPSS Inc.’s Predictive Analytics Software portfolio.