Power analysis — linear regression
Sample size or power for the F test of R² (or of an R² increase when only some predictors are tested), effect size f² = R² / (1 − R²).
Menu: Analyze → Power Analysis → Linear Regression…
When should I use it?
- Planning a multiple regression.
What are the assumptions?
- Fixed predictors; f² from earlier work (.02 small, .15 medium, .35 large).
How do I run it in StatLab?
- Give f² (or R²), the number of tested and total predictors, α and the target power.
How do I read the output?
| Output | Meaning |
|---|---|
| Total sample size / Actual power | Result of the search. |
Worked example
On the practice survey dataset (File → Sample dataset), this command — which the dialog's Paste button also produces — runs the procedure:
PROC pwRegression /mode=n /f2=0.15 /tested=1 /predictors=1 /alpha=0.05 /power=0.8.
Output tables: Power — linear multiple regression: fixed model, R² deviation from zero (a priori).
How would I report it in APA style?
A sample of 55 gives power .805 to detect the stated effect at α = .050.
Computed by StatLab's validated procedure code when this page was built. The example only illustrates the steps; choose variables that fit your own research question.
What are common mistakes?
- Setting 'tested' larger than the total number of predictors.