Aliasing happens because a fractional factorial leaves out runs, which makes the columns for some effects identical, so each estimate is really the sum of every effect in that alias chain. In a 4-run half fraction of three factors built with C = AB, the main effect of C is aliased with the AB interaction. The alias structure follows from the design generators and should be checked before running, so the effects most likely to matter are not tangled together.
Don't get excited about factor D yet. It's aliased with the AB interaction, and A and B are both big.
Also heard as
- alias
- alias structure
- alias chain
Do not mix it up with
- Confounding: Aliasing is the deliberate, known mixing built into a fractional design; confounding is the broader term, including mixing with blocks or with unplanned variables.
Related terms
Confounding
Mixing of two effects so that the data cannot separate them, whether by design or through an uncontrolled variable.
Resolution
Rating of a fractional factorial design, in Roman numerals, that states which effects are aliased with which.
Fractional factorial design
Experiment that runs a chosen fraction of all factor combinations, trading some aliased effects for far fewer runs.
Interaction
Situation where the effect of one factor depends on the level of another, so their combined effect is not simply additive.