ABSTRACT
Researchers require appropriate experimental design to test their hy potheses. Investigations should be conducted as efficiently as possible to detect even small differences among the treatments. Several designs are available. Thus researchers should consult a statistician before formulating the design and starting the experiment. Completely randomized design (CRD) is a basic design. It is also called a one-way analysis of variance model. Sometimes, it is also referred to as a one-way classification or one way layout model. In this design, the treatments are completely assigned at random to the experimental units without blocking. This is useful when the experimental units are uniform or homogeneous. When the experimental units are not uniform, one should use an alternate design.
