Course contentsShow
← Data Science
Lesson 465 of 2,14512. ANOVA and Multiple ComparisonsFree lesson

Interaction Effects

Understanding and interpreting interactions between two factors in a two-way design.

Interaction Effects

What you'll learn: How to recognize and interpret when the effect of one factor depends on the level of another factor in two-way ANOVA.

What Is an Interaction Effect?

In two-way ANOVA, you've learned about main effects — the average effect of each factor separately. But sometimes, two factors don't just add their effects together; they combine in unexpected ways. This is called an interaction effect.

An interaction occurs when the effect of one factor changes depending on the level of the other factor. The two factors work together in a way that's different from what you'd predict by simply adding their individual effects.

A Simple Analogy

Imagine studying how sunlight and water affect plant growth. You might find:

  • Sunlight alone helps plants grow (main effect of sunlight)
  • Water alone helps plants grow (main effect of water)

But here's the interaction: with no water, adding sunlight might actually harm the plant (it dries out). The effect of sunlight depends on how much water is present. That's an interaction — the two factors don't simply add up; they modify each other's effects.

Recognizing Interactions

In two-way ANOVA, you test for interactions before interpreting main effects. If a significant interaction exists, you can't simply say "Factor A increases the outcome" — you have to specify "Factor A increases the outcome when Factor B is at level X, but decreases it when Factor B is at level Y*."

Interaction plots are your best friend here: non-parallel lines between factor levels signal an interaction is likely present.

Key Takeaway: An interaction effect means one factor's impact changes depending on another factor's level — the two factors combine in non-additive ways, requiring careful interpretation beyond simple main effects.