Confounding occurs when you can’t distinguish the effects of certain factor interactions because of other potential factor effects. This most commonly happens when:
- Uncontrolled or ‘nuisance’ factors are affecting the results.
- Partial factorial designs are used.
For example, let’s look at a simple partial factorial design with confounding:
|Trial||Factor A||Factor B||Factor C|
In this experiment design, Fc = Fa * Fb (see Partial/Fractional Factorial Design if you need to revisit the notation used). That means that the effects of factor C on the experiment can’t be distinguished from the interactions between factor A and factor B.
Blocking can help limit confounding by distributing extra factors evenly across all experimental factors.