| Enter the Type I error rate, \(\alpha =\) | |
| Enter the population standard deviation, \(\sigma =\) | |
| Enter the population correlation between levels, \(\rho =\) | |
| How many levels (groups) do you have in your ANOVA? \(K=\) | |
| Number of iterations (bigger values produce better estimates, but take longer) |
| Power for all tests= | ||
| Sample size \(n=\) |
Provide the requested information for the test and the specific alternative hypothesis populations. This calculator assumes that the population standard deviation is the same for all levels, so it only needs to be entered once. It also assumes that the population correlation is the same for all pairs of levels, so it only needs to be entered once. You must enter the population mean for each level.
If you plan to run contrasts in addition to the ANOVA, then you should click on the Add a contrast test button. In the small table that appears set the contrast weights so that they sum to 0, define the directional nature of the alternative hypothesis, and set the criterion value. You can add as many contrast tests as you like.
When you click on the Calculate power button, the calculator will compute the power of each test and report the power of all tests combined. The latter value is provided in the Power for all tests text field. By entering value in that text field and clicking on the Calculate minimum sample size, the calculator will identify the smallest common sample size across all levels that produces at least the desired power.