(b) Determine the effects of the block and treatment terms on each response variable, by analysis of variance and by mixed modelling. For each significant relationship observed, state the nature of the effect – for example, does lactase raise or lower the level of lactose?

In each mixed-model analysis, the mean initial value of the response variable for each sheep can be added to the model. The difference between this mean and the corresponding individual initial value on each side of the udder in each sheep can also be added.