|
Two-Sample Equivalence TestTest |
In addition to the confidence interval (CI) for the difference (or ratio), Minitab also performs a hypothesis test. If you use the default method to calculate the CI, then both the test and the CI lead to the same conclusion about claiming equivalence.
The results for the hypothesis test include p-values for two separate null hypotheses:
If both null hypotheses are rejected, then the difference falls within your equivalence interval and you can claim that the means for the products are equivalent.
Example Output |
Test
Null hypothesis: Difference ≤ -0.5 or Difference ≥ 0.5 Alternative hypothesis: -0.5 < Difference < 0.5 α level: 0.05
Null Hypothesis DF T-Value P-Value Difference ≤ -0.5 12 1.8637 0.044 Difference ≥ 0.5 12 -3.0566 0.005
The greater of the two P-Values is 0.044. Can claim equivalence. |
Interpretation |
For the cat food data, you can reject both null hypotheses because the p-values for both tests (0.044 and 0.005) are less than the a level of 0.05. These results indicate that the difference is within your equivalence limits of -0.5 and 0.5. Thus you can claim that amount of protein in the discount cat food is equivalent to the amount of protein in the original formulation.