Main effect

average change when one factor goes from low to high

Effect(A) = ȳ(A = +) − ȳ(A = −)

The difference between the average response of all runs at the high setting of a factor and all runs at its low setting. On the main effects plot it is the slope of the line for that factor.

A large main effect means the factor matters, but if the factor is part of a significant interaction its effect also depends on the other factor.

Example

Average shrinkage 1.81 % at 210 °C and 1.37 % at 240 °C: the main effect of melt temperature is −0.44 %.

Calculate it

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

Related terms

• Interaction effect: the effect of one factor depends on another
• Pareto of standardised effects: effects ranked by size against a significance line
• Coded units (−1 / +1): factor settings written as low and high