A two-level factorial design is used to test the effects of changing more than one treatment factor at a time. Each factor is tested at both a high and low level.

Minitab's power and sample size capabilities allow you to examine how the following test properties affect each other:

·    power - the probability of being able to detect an effect of a given size.

·    replicates - the number of times each run is repeated.

·    effects - the difference in your variable that you want the test to be able to detect . This difference can be caused by a single factor alone (main effect), or a combination of factors (interaction).

·    center points - runs that are executed with all factors set at their middle settings.

If you enter values (or sets of values) for any three of these properties, Minitab will calculate the associated value(s) for the remaining property.

By default, all calculations are based on an a-level of 0.05. However, you can select any value between 0 and 1.

Researchers at a candy company are testing the effects of temperature and sugar concentration on the smoothness of the chocolate they use to cover their candy bars. They are interested in detecting a difference of 3.5 smoothness units or more. The estimated standard deviation is 1.9 units.