Power Analysis: Precision
The Precision dialog is available for every Power
Analysis procedure (except for Univariate Linear Regression), and provides options for estimating
the sample size based on specified confidence interval half-widths. The dialog is available when
Single power value or Grid power values are specified
for the Sample size estimate and when the Test
Direction setting is set to Nondirectional (two-sided
analysis).
Note: The Test Direction setting
does not apply to One-Way ANOVA.
when the Estimate sample size and Grid power values options are selected (click the Grid control to display the dialog).
- Confidence Interval Type(s)
- Select the appropriate confidence intervals types and whether continuity
correction is applied.Note: Confidence Interval Type(s) are available only for the Power Analysis proportion procedures (One-Sample Proportion, Related-Samples Binomial Test, and Independent-Samples Binomial Test).
- Specify the half-width of confidence interval
- Estimates the sample size based on the confidence
interval half-width value. Enter a value in the range 0 - 1. For the One-Sample Binomial test, the
value must be in the range 0 - 0.5.Note: Duplicate values are ignored.
- Click Add to add the specified half-width value to the list.
- Highlight an existing half-width value and click Change to update the value.
- Highlight an existing half-width value and click Remove to remove the value from the list.