Power Analyses by Distribution

Please note that at this point the G*Power documentation is still incomplete and preliminary. We take great care to avoid any errors, but of course we cannot guarantee that the documentation is error free.

G*Power provides dedicated power analysis options for a variety of frequently-used exact tests, F tests, t tests, χ² tests, and z tests. Simply click on the proper distribution name in the navigation bar on the left, then select the type of statistical test for which you want to perform a power analysis.

Note, however, that you are not limited to these particular tests because G*Power also offers power analyses for generic binomial tests, F tests, t tests, χ² tests, and z tests for which the noncentrality parameter of the distribution under H₁ may be entered directly. In this way, you are provided with a flexible tool for computing the power of basically any statistical test that uses binomial, F, t, χ², or z reference distributions. The generic test option is always the last item in the list of tests for one particular distribution.

Alternatively, you may use the design-based approach : Simply select (a) the parameter class the statistical test refers to (i.e., correlations, means, proportions, variances) and (b) the design of the study (e.g., number of groups, independent vs. dependent samples, etc.).