Answers are grounded in this paper and cite their sources.
Monte Carlo estimates are obtained of the small sample power of the classical F test and three competing rank tests for detecting two forms of treatment shift alternatives. These results were computed under Scheffé's mixed model for the case of one observation per cell using several variance-covariance matrices. Some small sample null distributions, empirical efficiencies, and robustness computations are also given.
No takes yet. Share an insight, caveat, or question.
R.O. Gilbert (1972) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: