Three diets and beef from cows

This diagram extends the earlier simulation involving two diets (factor levels) and matched pairs to an experiment with three factor levels in which the experimental units are grouped into matched groups of three.

As before, click Accumulate then repeat the completely randomised experiment several times, noting the accuracy of the estimated differences between the pairs of diets.

Select Grouped by calf body weight to collect the calves into groups of three with similar initial body weights. Again repeat the experiment several times in this randomised block design and note that the treatment differences are much more accurately estimated.