Supplementing Interaction: When no Follow-up is Needed (1 of 2)

---

A significant interaction indicates that the effect of one variable differs depending on the level of another variable. Often, this is exactly the information that a researcher needs, and no follow-up analyses are necessary. For example, consider a hypothetical experiment designed to test whether hyperactive and non-hyperactive children are affected differently by the presence of distracting stimuli. The experiment has two factors: child's classification (hyperactive or non-hyperactive) and distraction (distraction present and distraction absent). A graph of the means for the four conditions is shown below.



The graph shows that the effect of distraction is greater for the hyperactive children than it is for the non-hyperactive children. If the interaction were significant, then the researcher would be able to conclude that distracting stimuli are more disruptive to the hyperactive children than they are to the non-hyperactive children.