What value do I use for standard deviation in a power and sample size analysis?

The standard deviation (also called sigma or σ) that you use in a power and sample size analysis depends on whether you already collected the data.

If you have not collected the data, use an estimate of the standard deviation for the population. Base your estimate on related research, design specifications, pilot studies, subject-matter knowledge, or similar information.

If you already collected and analyzed the data, use a value from the analysis results as shown in the following table.

Type of analysis that you performed	Value to use for standard deviation
1-sample t or 1-sample Z	Standard deviation of the sample
2-sample t	Pooled standard deviation
Paired t	Standard deviation of the paired differences
One way ANOVA, 2-level factorial design, or Plackett-Burman design	Mean square root of the mean square of the error (MSE)

Note

If you would like to specify the difference (effects) in standardized (sigma) units, enter 1 in Standard deviation.