Extended tables of the distributions of rank statistics for treatment versus control in randomized block designs
Title:
Extended tables of the distributions of rank statistics for treatment versus control in randomized block designs
Author:
Odeh, Robert E.
Appeared in:
Communications in statistics
Paging:
Volume 6 (1977) nr. 2 pages 101-113
Year:
1977
Contents:
In this paper we give the distribution of rank statistics for testing treatment versus control in a randomized block design. The data are given in the form of a two-way layout of n rows and k columns. The rows represent blocks and the columns represent treatments. We assume that treatment 1 is the control or standard. Denote by Xij (i=l,…, n; j=l,…, k) the observation in the ith row and jth column. Let [image omitted] , where rij is the rank of Xij within the ith block. Define R* = max {(Rj-R1 ): j=2,…, k} and define R** = max{∣ Rj -R1∣: j=2, …, k}. Tables are given for the upper tail probabilities for the null distribution of R* and R** for k=2(l)5, n=2(l)8 and for k=6, n=2(l)6.