Factor Levels

Factor levels are the number of conditions, or levels, that a given factor will be run in the experiment. An example of a factor to be run at two levels is when the factor temperature is set at either 60 deg or 85 deg for the runs in an experiment. Nonlinear designs , and Response Surface Analysis, require more than 2 levels.

The choice of number of levels influences the availability of specific nonlinear designs such as the Central Composite and Box-Behnken .

The levels for quantitative factors are usually based on the experimental range and on a need to evaluate these nonlinear effects. Those forqualitative factors are usually set by the categories inherent in the experiment (for example, Supplier A and Supplier B).

Learn more about the Regression tools in Six Sigma Demystified (2011, McGraw-Hill) by Paul Keller, in his online Regression short course (only $99), or his online Black Belt certification training course ($875).