Abstract: | Consider testing the hypothesis of no treatment effects against a postulated ranking of the m treatments, given data from n Complete Blocks. A suitable test statistic is the weighted average rank correlation w = σbQiCi where Ci is the correlation between the postulated ranking and the ranking observed within the ith block, Qi is the rank of the ith block with respect to credibility, and the bi's are weights such that 0 ≦ b1 ≦ … ≦ bn. In this paper we introduce some simple statistics: the first extends the signed-rank statistic to m ≦ 3, the second uses a simple measure of correlation based on the antirank, and the third a statistic based on Spearman's footrule. Tables for critical values are provided and the normal approximation is investigated. |