What is it about?
Although NHST is confusing and often misunderstood, researchers, as well as journal editors and reviewers, know intuitively that null hypothesis testing serves a useful purpose. It helps to screen out experimental effects that are so small that they are not likely to replicate or lead to more refined results and it encourages individual researchers to use adequate power for their studies, and not clutter the inboxes of journal editors with unreliable studies. Instead, tiny effects and null effects should be posted in searchable online repositories to avoid duplicating efforts and to enable more wide-ranging meta-analyses.
Featured Image
Why is it important?
My article presents a fairly strong and rather rare defense of the method most widely used by psychologists to evaluate the reliability of their studies. I take a balanced and nuanced approach to the topic and use numerical examples to support my arguments.
Perspectives
My hope is that this article will clarify the issues surrounding null hypothesis significance testing for those who need what is mostly an accessible, plain language assessment of this very important research issue.
Dr. Barry Howard Cohen
New York University
Read the Original
This page is a summary of: Why the resistance to statistical innovations? A comment on Sharpe (2013)., Psychological Methods, January 2017, American Psychological Association (APA),
DOI: 10.1037/met0000058.
You can read the full text:
Contributors
The following have contributed to this page
