Profile analysis for randomized complete block experiments

We consider the use of standard univariate and multivariate methods for profile analysis of randomized complete block experiments. Although the analysis for the case where the block time interaction is included in the model parallels that used for factorial experiments, situations where such interaction is not present may not be handled in the same way. We identify hypotheses for which the standard analysis may be applied, as well as those for which some adaptation is required. We also indicate how to implement such analyses via existing computer software.