Published online by Cambridge University Press: 03 October 2008
It is suggested that quite a small computer, in association with a minimal program for the analysis of variance, can be used to calculate quantities of use to the agronomic research worker beyond what he usually obtains. For example, without further programs it is possible to calculate an analysis of covariance, in which adjustment is made for some disturbing factor. It is also possible to find how much of the error variance arises from a particular plot, and to deal with situations in which data are incomplete or the yields from two plots have become mixed.