ABSTRACT
Chapter 13 discussed in detail some of the classical statistical analysis methods able to process data from experiments influenced by several factors. Very conveniently, in Chapter 13 the data happened to be such that these methods could be applied. Measurements were available for all needed factor interactions. However, this does not happen automatically in every experiment involving data collection. In order for the experiment to provide the data necessary for the analysis, the experiment needs to be designed. The design of the experiment is a crucial but often neglected phase in microarray experiments. A designed experiment is a test or a series of tests in which a researcher makes purposeful changes to the input variables of a process or a system such that one may observe and identify the reasons for changes in the output response [315]. If the experiments are not designed properly, no analysis method will be able to obtain valid conclusions. It is very important to provide data for a proper comparison for every major source of variation.
