ABSTRACT
Hochberg (1975) derived a two-stage multiple comparison procedure that addresses an issue of fundamental importance: How many additional observations must be sampled, if any, to achieve some desired level of precision when estimating a collection of linear contrasts. In particular, given that the length of a confidence interval should not exceed some specified value, how many additional observations are required to achieve this goal? A related problem is how to control the simultaneous probability coverage, or the familywise error rate (which is defined as the probability of at least one Type-I error among a collection of hypotheses being tested) in the event that additional observations can be acquired.
