# Enter your data for Power and Sample Size for General Full Factorial Design

Stat > Power and Sample Size > General Full Factorial Design

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

1. In Number of levels for each factor in the model, enter 2 to 15 values separated by spaces. Each value is the number of levels for a factor in the experiment.
2. Specify values for two 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 experimental run. For example, to calculate the effect of setting up each experimental run 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.
• Values of the maximum difference between main effect means: Enter one or more values to specify the mean difference that you want to detect between the low and high levels of a factor. The calculations use the factor that has the most levels to produce calculations that are conservative for other factors. Usually, you enter the smallest difference that has practical consequences. 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 probability that the test correctly detects the maximum difference between the level means of a single factor. Common values are 0.8 and 0.9. For example, analysts enter 0.9 for a 90% chance that the test will detect a practical difference in the strength between the smallest and largest means of a factor.
3. 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.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy