Power Tables for the One-Sided F Test Used to Determine Differences between Two Population Variances.

Abstract

Abstract : When comparing two population variances based upon sample data, one often uses a common statistical hypothesis test known as the F test. This test uses the ratio of the sample variances as a test statistic to determine if the population variances differ. The ratio is compared with a predetermined critical value that is dependent upon the sample sizes and the significance level of the test. The power of the test-that is, the probability of exceeding this critical value can be determined a priori if one assumes to know the actual ratio of population variances. This report details the use of the F test and gives extensive tables of power for various significance levels, population variances ratios, and sample sizes under 100. It is hoped that this report will allow researchers to make wise choices of sample size in conducting future experiments of this type.