Overview for Equivalence Test with Paired Data

Use Equivalence Test with Paired Data to evaluate whether the mean of a test population is equivalent to the mean of a reference population when you have paired (dependent) observations.

When you use an equivalence test with paired data, you must specify a range of values that are "close enough" to be considered equivalent to the reference mean. This equivalence interval, also called the zone of equivalence, is based on your knowledge of the product or process and should be determined before you perform the test. The analysis then determines whether you have enough evidence to claim that the difference (or ratio) between the population means is within the equivalence interval.

For example, an analyst wants to test whether a new glucose meter is equivalent to a currently approved glucose meter. The analyst will consider the new meter equivalent if it produces mean glucose readings within ±20% of the current meter. The analyst measures blood glucose twice in each person, once with the new (test) meter and once with the currently approved (reference) meter. The measurements are paired data (dependent observations) for each person. If the confidence interval for the ratio of the test mean and the reference mean is within the equivalence interval of 0.80 and 1.20, the two glucose meters are equivalent.

You can also use the equivalence test with paired data to perform superiority tests and inferiority tests, to evaluate whether the mean of a test population is greater than or less than the mean of a reference population.