5 The randomised complete block design

Variability from a nuisance factor, a factor that may affect the response but is of no interest in itself, can obscure the treatment comparisons. How it is handled depends on what is known about it.

(a)

Unknown and uncontrolled: randomisation, which spreads its effect evenly over the treatments in the long run (Section 1).

(b)

Known but uncontrollable, though its value can be observed on each run: the analysis of covariance, which adjusts the response for the measured value.

(c)

Known and controllable: blocking, the subject of this section.

Questions on this section

Stuck on something here? Ask below and it stays attached to this topic.