Effect sizes and "what if" analyses as supplements to statistical significance tests

As shortcomings of traditional statistical significance testing have been highlighted, calls have been heard within the research community for alternative or supplemental analyses that reflect the practical significance of test results and the role of sample size in the interpretation of results. Effect size reporting, which is designed to reflect on the practical value of results, is now required by 23 journals, including the flagship journals of the American Counseling Association, the Council for Exceptional Children, and several American Psychological Association journals. "What-if" analyses, though less well established, offer a method for reflecting on the role of sample size in the decision to accept or reject the null hypothesis. Explanations of these methods and guidelines for their use and interpretation are offered.

Overview