# Enter your data for Power and Sample Size for 2-Level Factorial Design

Stat > Power and Sample Size > 2-Level Factorial Design

Complete the following steps to specify the data for the power and sample size calculation.

1. In Number of factors, enter the number of variables that you plan to control in the experiment. The number must be between 2 and 15.
2. In Number of corner points, enter the number of experimental runs where the factors are set at their low or high levels. This number is for one replicate of the experiment. Valid numbers depend on the number of factors. For a table of valid combinations, go to Available 2-level factorial designs.
3. Specify values for three of the following power function variables. Leave the variable that you want to calculate blank.
• Replicates: Enter one or more positive integers to specify the number of times that you set up each corner point. For example, to calculate the effect of setting up each corner point twice, enter 2. To assess the effect of different numbers of replicates, enter multiple values. More replicates give the experiment more power to detect an effect and can increase the precision of predictions.

• Effects: Enter one or more values to specify the difference in the mean response that you want to detect between the low and high levels of a factor. Usually, you enter the smallest effect that has practical consequences for your application. For example, enter 5 if this difference between means is important to detect, but differences smaller than 5 are less important.

• Power values: Enter one or more values to specify the power that you want the experiment to have. Common values are 0.8 and 0.9. For example, enter 0.9 for a 90% chance that the test will detect a practical difference between the high and low factor settings.

• Number of center points per block: Enter one or more non-negative integers to specify the number of experimental runs per block where the factors are at their middle settings. To change the number of blocks, click Design.

Common numbers of center points in a design usually depend on other considerations besides power. For example, you can use center points to check for curved relationships between the factors and the response. The most general advice is to use at least 3 center points in the design. For example, 2 center points per block in an experiment with 2 blocks produces a design with 4 total center points.

4. In Standard deviation, enter the standard deviation of the response measurements at replicated experimental runs. Usually, you estimate this value from related research, pilot studies, or subject-matter knowledge. If you already performed an analysis in Minitab that produced an ANOVA table, you can use the square root of the adjusted mean square for error. You can also enter 1. When you enter 1, the effect sizes are multipliers of the standard deviation instead of units of the response variable. For example, if you specify an effect size of 2 and a standard deviation of 1, then the calculations are for an effect that is the size of 2 standard deviations.
###### Note

Because you specify the number of center points per block, the total size of the experimental design also depends on the number of blocks. To specify more than one block, click Design.

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