Comparisons of means of many populations

Abstract

Testing hypotheses on the equality of means is very common. Methods of comparing simultaneously many populations with regard to their means are less popular than the methods of comparing two populations. For this reason, the paper deals only with the latter case. Several tests used for the verification of hypotheses on the equality o f many means are described. If such a hypothesis is rejected with these tests, we can find the so-called homogenous groups. An example is given showing how to use the tests and how to interpret the results obtained.