Free Online Lean Six Sigma Encyclopedia

English |  Español |  Français |  Português |  Deutsch |  中文

Balanced Factorial Design

Go Back


A balanced design is one in which the same number of measurements or Replications is assigned to all treatment combinations in the experiment. Thus, there are an equal number of observations in each cell.

A balanced design is desirable for its orthogonality property and because it tends to be robust to small violations of assumptions necessary for the analysis, such as those of normality and equal Variances. However, sometimes balance may be impossible to achieve due to shortage of Experimental Units, missing data or other constraints.