Hide/Show Apps

Mean homogeneity tests under the violation of the assumptions

Sakmar, Elçin
In social, educational and health sciences, obtained data are generally non-normal and heterogeneous although many methods expect one or more assumptions to be satisfied. However, there is still a need for an extensive Monte Carlo comparison and a new way for testing the means for populations that have both non-normality and heterogeneity. So, the first aim of the current study is to compare several means for both under non-normality and variance heterogeneity. The second aim is to compare several means that are normally distributed with heterogeneity of variance. The B = 1000 bootstrap tests are performed for