Fractional Factorial ANOVA

In designed experiments, a fractional factorial design (as opposed to Factorial ANOVA) is one where not all factor combinations are observed and represented in the data set, but instead, only specific factor combinations are observed so as to allow for the testing of specific hypotheses (e.g., main effects, two-way interactions). Fractional factorial ANOVA designs are usually applied when the number of factor combinations is very large and/or prohibitively expensive to observe.