Aliasing (confounding)

effects that cannot be separated

In a fractional design some columns of the design are identical, so the effects behind them are estimated together. If A is aliased with BCD, the calculated effect is the sum of both; usually the higher-order interaction is assumed to be negligible.

A full factorial has no aliasing.

Calculate it

DOE: 2-Level Factorial Design: 2^k factorial: run plan, effects, Pareto, interaction plots and best settings.

Related terms

• Fractional factorial design: a planned subset of a full factorial