Ruediger Vollandt, Manfred Horn (University of Jena, Germany)

Sample size determination for multiple many-one and pairwise comparisons of proportions

We address the problem of sample size determination in many-one and pairwise multiple comparisons of proportions which are arcsin-root transformed. For one- and two-sided many-one comparisons, we provide the least favorable configuration which minimizes the all-pairs power. Corresponding explicit sample size formulas are given, also for the case of prior knowledge on the underlying success probabilities. The solutions for pairwise comparisons are restricted to the case *k* = 3.

References:

- Hayter, A.J. and Liu, W. (1990)."Power asessment for tests of equalityof several proportions," Commun. Statist. - Theory Meth., 19, 19-30.
- Hayter, A.J. and Lui, W. (1992). "A method of power assessment for tests comparing several treatments with a control," Commun. Statist.-Theory Meth.., 21, 1871-1889.
- Horn, M. and Vollandt, R. (1998)."Sample sizes for comparisons of k treatments with a control based on different definitions of the power," Biometrical Journal,40, 589-612.
- Liu, W. (1996)."On some single-stage, step-down and step-up procedures for comparing three normal means," Computational Statistics & Data Analysis, 21, 215-227.