Understanding the Minimum Effect of Interest

The Minimum Effect of Interest (MEI) is a business term meaning the smallest effect that has business value. It is usually used to determine the minimum sample size needed in an experiment. When the data is collected, the effect size is calculated, and if the effect size is smaller than the MEI, the result is considered to be not practically significant.

Calculating the Minimum Effect of Interest

The formula to calculate the minimum sample size is:

$\mathbf{N} = \frac{2(Z_{1-\alpha/2} + Z_{1-\beta})^{2}{CV}^{2}}{{MEI}^{2}}$

Where:

• $N$: the minimum sample size
• $Z_{1-\alpha/2}$: the Z-score for the desired confidence level
• $Z_{1-\beta}$: the Z-score for the desired power
• $CV$: the coefficient of variation
• $MEI$: the minimum effect of interest, it’s usually set to 0.02 or 0.05

Difference between Minimum Effect of Interest and Minimum Detectable Effect

There’s another famous term in statistics called Minimum Detectable Effect (MDE), which is the smallest effect that can be detected with a given sample size. MEI and MDE are similar but generally in the opposite direction. The minimal sample size is the output of the MEI, but the real sample size is the input of the MDE.