ABSTRACT
The randomized complete block design (RCBD) is used for studying the effect of treatment in the presence of a nuisance factor, which is not of primary interest in the study. Ordinarily, the study that is modeled as RCBD would have been treated as a completely randomized design (CRD). The nuisance factor (usually called the block), is designed into the experiment to ensure that if its effect is large, it would not inflate the size of the error sum of squares, carry through to the ANOVA, and thus depress the sensitivity of the ANOVA to detect the significance of the treatment effect. For instance, an investigation may be such that it stretches over different shifts of workers of differing skills. The shifts in such a study could be treated as blocks. Alternatively, treatments may be such that they are carried out by different groups of workers with differing expertise and experience. The groups could be treated as blocks and the investigation carried out under RCBD. We discuss the RCBD in the next few sections. The RCBD is important because after initially being applied to selected agricultural experiments, it is now used extensively in agricultural, medical and industrial experiments. For example, in agricultural experiments, there are often different fertilizers tested on various types of soils to determine which one gives the best yield for a crop. The types of soils may be clay-like, loamy, humus, or sandy. To design an experiment to study the yields of soybean under five fertilizer treatments on soils that are clay-like, loamy, humus, or sandy, the soil type is treated as the nuisance factor in the experiment and would be designed into the experiment as the block.
