The POWER Procedure
- Overview
-
Getting Started
-
Syntax
PROC POWER Statement LOGISTIC Statement MULTREG Statement ONECORR Statement ONESAMPLEFREQ Statement ONESAMPLEMEANS Statement ONEWAYANOVA Statement PAIREDFREQ Statement PAIREDMEANS Statement PLOT Statement TWOSAMPLEFREQ Statement TWOSAMPLEMEANS Statement TWOSAMPLESURVIVAL Statement TWOSAMPLEWILCOXON Statement
-
Details
Overview of Power Concepts Summary of Analyses Specifying Value Lists in Analysis Statements Sample Size Adjustment Options Error and Information Output Displayed Output ODS Table Names Computational Resources Computational Methods and Formulas ODS Graphics ODS Styles Suitable for Use with PROC POWER
-
Examples
One-Way ANOVA The Sawtooth Power Function in Proportion Analyses Simple AB/BA Crossover Designs Noninferiority Test with Lognormal Data Multiple Regression and Correlation Comparing Two Survival Curves Confidence Interval Precision Customizing Plots Binary Logistic Regression with Independent Predictors Wilcoxon-Mann-Whitney Test
- References
Analyses in the ONEWAYANOVA Statement
One-Degree-of-Freedom Contrast (TEST=CONTRAST)
The hypotheses are
 |
 |
|||
 |
 |
where 
is the number of groups, 
are the contrast coefficients, and 
is the null contrast value.
The test is the usual 
test for a contrast in one-way ANOVA. It assumes normal data with common group variances and requires 
and 
.
O’Brien and Muller (1993, Section 8.2.3.2) give the exact power as
 |
where
 |
Overall F Test (TEST=OVERALL)
The hypotheses are
|
 |
|||
|
 |
where is the number of groups.
The test is the usual overall test for equality of means in one-way ANOVA. It assumes normal data with common group variances and requires and .
O’Brien and Muller (1993, Section 8.2.3.1) give the exact power as
 |
where the noncentrality is
 |
and
 |