Factorial Design
Factorial designs are used for experiments that test the effect of more than one treatment. This type of design permits an assessment of interactions among the treatments. A factor (treatment) can have more than one level, for example the levels could be different doses of the same drug. A factor can also be defined as the presence or absence of a single drug. The levels of a factor would NOT be drug A and drug B (assuming neither is a placebo). An essential feature of this design is that the factors are varied such that some groups will receive more than one treatment. Using this design we can test if the combination of treatments is better (or worse) than individual treatments.
Factorial designs are very efficient when the goal of the study is to compare treatments that are known not to interact. Because of this we need firm biological knowledge to support the purpose for the trial. If the goal is to estimate the interaction between treatments then the factorial design, while no longer efficient, is the only type of trial design that can be used.
When conducting a factorial design we need to keep the following rules in mind:
- Treatments must be amenable to being administered in combination without changing dosage in the presence of each other.
- It must be ethically acceptable not to administer the individual treatments, or administer them at lower doses as the case may be.
- There must be genuine interest in learning about treatment combinations.
- Treatments acting through the same mechanism are not appropriate for factorial designs.
-- ErinEsp - 01 Feb 2010