Data considerations for Equivalence Test with Paired Data

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret the results.

The sample data should not be severely skewed, and the sample size should be greater than 20

If your sample sizes are greater than 20 and the underlying distribution is unimodal and continuous, the test performs appropriately even if data are mildly skewed. If your sample sizes are less than 20, you should graph the data to check for skewness and unusual observations. If the data are severely skewed or have many unusual observations, use caution when you interpret the results.

The sample data should be selected randomly

In statistics, random samples are used to make generalizations, or inferences, about a population. If your data were not collected randomly, your results may not represent the population. For more information, go to Randomness in samples of data.

You should have a set of paired (dependent) observations, such as two readings of blood pressure taken from the same patient

If you have two independent samples (such as samples from two different groups of patients in a parallel study design), use 2-Sample Equivalence Test. For more information, go to How are dependent and independent samples different?.

Determine an appropriate sample size

Your sample should be large enough that the test has sufficient power to demonstrate equivalence when it is true. If the equivalence test has low power, you may mistakenly conclude that the difference (or ratio) is not within your equivalence limits when it actually is. To determine the appropriate sample size for your equivalence test, go to Power and Sample Size for Equivalence Test with Paired Data.