No abstract

tion of the number of subjects than would otherwise be required for a completely crossed design, by confounding some of the effects with higher order interactions (Kirk, 1968). Therefore, the number of hypotheses tested would be smaller than what would ordinarily be tested in a complete design, and the likelihood of a Type I error due to excessive multiple testing is diminished.Either, or any combination, of the above three techniques would be appropriate strategies for controlling the Type I error rate for factorial experiments, while also permitting the researcher to benefit from the efficiency of factorial experimentation.REFERENCES

No abstract