Go to the GraphPad home page
  GraphPad Home Library Biostatistics -- specialized areas Testing statistical hypotheses of equivalence.

In some kinds of experiments, your goal is to show that two treatments lead to results that (while not identical) are close enough to be equivalent. You might think that it is enough to show that the difference between the two groups is "not statistically significant", but in fact proving equivalence is more complicated than that.

Free GraphPad resources

Statistical tests for equivalence   It isn't hard to interpret experiments testing for equivalence. Look at the confidence intervals, use plenty of common sense, and don't bother with P values or statements of statistical significance. This short article explains how to do it.

Links to other sites

Bioequivalence   One application that uses the ideas of statistically testing for equivalence is when you test whether two formulations of a drug have equivalent biological action. The FDA explains in this document.

The ideas behind equivalence testing   A short set of slides from Walter Hauck (Thomas Jefferson University) explaining the ideas behind equivalence testing.

Equivalence testing   Slides by Brad Warner, using military rather than biological examples, that explain the principles.

Absence of evidence is not evidence of absence   Altman and Bland [BMJ 1995;311:485 (19 August)]  concisely explain the problems of interpreting "negative" studies.

A Bayesian looks at equivalence   The first half of these PowerPoint slides explains the general issue of testing for equivalence and how statistical hypothesis testing can be adapted to test for equivalence. The second half explains a Bayesian perspective.

Recommended books

Testing Statistical Hypotheses of Equivalence
by Stefan Wellek and Stefan Welleck, Chapman & Hall, 15 November, 2002.
ISBN 1584881607. List price: US$89.95.
Buy from amazon.com for US$72.86

Detailed explanations of both the principles behind equivalence testing, and the nitty gritty of how to apply these principles to different kinds of tests. Too much detail for someone just getting started. View the table of contents.