ABSTRACT
We have become accustomed, in the last four chapters, to calculating an analysis of variance. From the analysis of variance we obtain an estimate, s
, of the random variance between plots within blocks which we then use to calculate SEs of differences between the mean values for different treatment combinations. This is considerably more complex than our original use of, for example, the t-test to compare the difference between the means of two distinct samples with a hypothetical zero difference. Consequently, some of the assumptions that appeared intuitively acceptable and even obvious for the original simple situation need restating at this stage of the book, and need to be considered each time an analysis of variance is carried out.
