A group sequential procedure for comparing several treatments with a control.
Journal of Applied Statistics, 23, (4), . (doi:10.1080/02664769624116).
Full text not available from this repository.
The purpose of this paper is to provide some group sequential procedures for comparing several treatments with a control. These procedures are the generalizations to the sequential sampling setting of the single-step procedure of Dunnett and the step-down procedure of Hochberg and Tamhane. When the responses are independent normal random variables with a common known variance, the critical values for specified alpha-levels are tabulated. The situation where the responses are binary random variables is also discussed.
Actions (login required)