Suppose the experimenters are interested in how weight gain in rats is affected by different sources of protein: Beef, Cereal and Egg.

Suppose there are $30$ heterogeneous rats in terms of initial weight and there are two groups of rats identified prior to the experimentation: rats with low initial weight and high initial weight. However within each group, rats are homogeneous in terms of appetite size. Each treatment has $10$ replicates. This experiment will be a randomized block design.

How can I tabulated the design?

Here, different sources of protein (Beef, Cereal and Egg), are the treatments and low initial weight and high initial weight are two blocks. But I can't set the number of replicates so that each treatment has $10$ replicates. $$\text{Treatment}$$ $$ \begin{array}{c|ccc} \text{Blocks}&\text{Beef} &\text{Cereal}&\text{Egg}\\ \hline\\ \text{low initial weight}&y_{11}&y_{21}&y_{31}\\ \text{high initial weight}&y_{12}&y_{22}&y_{32}\\ \end{array} $$

Or will it follow randomized complete block design with multiple observations per cell?

Am I missing any information such that how many observations are there in a block in each treatment?

Can you provide me the data layout?

  • 1
    $\begingroup$ I don't know if I'm really understanding your question correctly, but either you display the number of observations (i.e., 5), or you tabulate your data. In the latter case, you probably want to display the mean across all five observations per cell replacing $y_{11}$ with $\bar{y}_{11}$ $\endgroup$ – hplieninger Aug 16 '13 at 10:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.