Data - Equivalence Test with Paired Data
main topic        

Data must be paired, or dependent observations. For example, suppose that you want to determine whether 2 blades are equally good at cutting  through leather. Because pieces of leather vary in grain and toughness, you decide to take a pair of measurements from each piece, one using the test blade and one using the reference blade. Because both measurements are affected by the toughness of the particular piece, they are dependent.

Measurements for the test condition are in one column and measurements for the reference condition are in another column. Each row contains a pair of dependent observations.  

If either measurement in a row is missing, Minitab omits that row from the calculations.

For the log transformation option, all observations must be greater than zero.