ABSTRACT
This chapter outlines several types of scientific studies in focusing in particular on designed experiments. Designed experiments formally randomize over extraneous factors which might influence the response, while controlling assignment to groups. The purpose of a designed experiment is to provide a logical framework for comparing groups in terms of some response of interest. Factors under study in an experiment may act in combination, or interact, in their effect on a response of interest. A designed experiment consists of the following steps: determine the key questions which drive the experiment; choose a treatment structure; select a design structure; conduct the experiment; and gather measurements for each experimental unit. Design structure is the most important part of the experiment to understand because it is so often overlooked in the rush to examine the treatment structure. As confidence comes with comprehension of the experimental design, data analysis can become more sophisticated.
