Effect Size as a Complementally Statistical Procedure of Testing Hypotheses

Abstract

This study aimed at presenting the concept of “effect size” as a complementary statistical procedure of testing hypotheses. The study included numerical analysis to prove that the magnitude of effect size is less influenced by the sample size than the statistical procedures used in testing hypotheses. Furthermore, the study displayed some statistical measures used to show the magnitude of effect size in the case of some statistical procedures. Specifically, t- test either in the case of independent samples or related samples, F- test in the case of university statistics? Pearson correlation coefficient? And simple regression. Also the study attempted to clarify the relationship between effect size and statistical power and indicated to the positive relationship between both of them.