F- and H-Test Assumptions Revisited

Abstract

The effects of some restraints not included in the classical assumptions of the F- and H-test (e.g., correlation of mean and sample size) were examined in a simulation design of 1000 samples per condition. Also simulated was a situation in which two assumptions were not met simultaneously. The major conclusions were: H was not an appropriate alternative for F with samples of N? 20; in all cases of unequal variances combined with unequal sample sizes H should be applied; and neither H nor F should be applied if more than one assumption of either test is not met.