If you specify covariates in the model (whether continuous or categorical), then two adjustments are made in order to compute
approximate power in the presence of the covariates. Let denote the number of covariates (counting dummy variables for categorical covariates individually). In other words,
is the total degrees of freedom used by the covariates. The adjustments are as follows:
The error degrees of freedom decrease by .
The error standard deviation shrinks by a factor of
(if the CORRXY= option to specify the correlation
between covariates and response) or
(if the PROPVARREDUCTION= option is used to specify the proportional reduction in total
incurred by the covariates). Let
represent the updated value of
.
As a result of these changes, the power is computed as
![]() |
where is calculated using
rather than
:
![]() |