ABSTRACT

One purpose of factorial designs is to use the same data to test the ef fect of more than one factor. If there are

a

levels of one factor and

b

levels of a second factor in a two-factor experiment, the number of cells in the matrix of treatment combinations is

ab

and each cell represents a distinctive treatment. Many tests are possible: The data can be used to test separately the ef fects of factor A and factor B, or we can test the ef fect of either factor within any single level of the other factor. Thus, by using the same data for various comparisons, we can make more complete use of our data than in a single-factor design.