Visualizing Interactions
What you'll learn: You'll discover how to use interaction plots to spot and understand when two factors combine in unexpected ways to influence your outcome.
Why Interaction Plots Matter
In Two-Way ANOVA, you've learned that interaction effects occur when the impact of one factor depends on the level of another factor. But detecting and interpreting these interactions from numbers alone is difficult. Interaction plots make these non-additive effects visible at a glance.
What Is an Interaction Plot?
An interaction plot displays group means on the y-axis, with one factor's levels on the x-axis, and separate lines representing each level of the second factor. If the lines are roughly parallel, there's likely no interaction—the effects are additive. If the lines cross or diverge significantly, you've got an interaction: the two factors don't simply add up; they modify each other's effects.
Analogy: Coffee and Sugar
Imagine studying alertness based on coffee (caffeinated vs. decaf) and sugar (with vs. without). If both factors work independently (no interaction), adding sugar increases alertness by the same amount whether you drink caffeinated or decaf coffee—parallel lines. But if sugar only boosts alertness in caffeinated coffee and does nothing in decaf, the lines would diverge or cross—revealing an interaction.
Reading the Plot
- Parallel lines → No interaction; effects are independent
- Crossing or converging lines → Interaction present; one factor's effect changes depending on the other
- Steeper slopes at different levels → The magnitude of one factor's effect varies by the other
Key Takeaway: Interaction plots turn complex factorial relationships into clear visual patterns—parallel lines mean additive effects, while crossing or diverging lines reveal interactions where factors influence each other.