What is it about?
Researchers often ask questions about whether an effect is small enough to be considered unimportant. For example, is the difference in productivity between working at home and working in the office negligible? Standard statistical methods are primarily designed to detect relationships among variables (e.g., mean differences), but questions regarding negligible relationships require a special statistical framework called negligible effect (equivalence) testing. We provide an overview of this framework in this tutorial.
Featured Image
Photo by Tolga Ulkan on Unsplash
Why is it important?
We provide a step-by-step walk through of scenarios where negligible effect tests are relevant, how they work, and how to conduct them. This allows researchers to recognize questions where negligible effect tests are applicable and gives them guidance on how to conduct the tests themselves.
Perspectives
It was a pleasure to bring together so much of the rich literature on negligible effect testing into one place. Although researchers are often interested in demonstrating negligible associations among variables, they often do so inappropriately using traditional hypothesis testing methods that are designed to detect the presence of relationships. We hope this tutorial provides researchers with the tools necessary to rigorously examine questions regarding negligible effects in their own work.
Nataly Beribisky
Read the Original
This page is a summary of: A primer on equivalence (negligible effect) testing., Psychological Methods, February 2026, American Psychological Association (APA),
DOI: 10.1037/met0000800.
You can read the full text:
Contributors
The following have contributed to this page
