Select the analysis options for Equivalence Test with Paired Data

Stat > Equivalence Tests > Paired > Options

Risk level of claiming equivalence when it is not true (alpha)

Specify the risk level (denoted as α or alpha) for the test. Usually, an α of 0.05 works well. An α of 0.05 indicates a 5% risk of rejecting the null hypothesis when it is actually true (type I error).

The value of α is often set by industry guidelines. If you use a smaller value for α, such as 0.01, you can reduce the risk of a type 1 error. For example, if you perform an equivalence test using the default hypotheses, using a smaller alpha reduces the risk of claiming equivalence when it is not actually true. However, a smaller α also increases the risk that the test is overly conservative and does not allow you to claim equivalence when it is actually true.

The α value also sets the confidence level for the confidence interval. By default, the confidence level is (1 – α) × 100%. For example, if α is 0.05, Minitab calculates a 95% confidence interval.

Use (1 - 2 alpha) x 100% confidence interval

Select to use an alternative method to calculate the confidence interval for the difference (or the ratio) between the population means. For a specific value of α (alpha), the confidence level for the alternative method is always smaller than that of the default method. For example, if α is 0.05, then the confidence level is 90% for the alternative method and 95% for the default method.

The alternative method is sometimes requested by regulatory guidelines. Although the alternative method often produces the same upper confidence limit or lower confidence limit as the default method, the interval it produces may be more conservative (or more liberal) than the corresponding α-level test. The default method is generally more powerful and corresponds more closely with the results of the α-level test.

By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy