Assigning Properties

Properties	Description
Solve For
requests a solution for power or total sample size.
Analysis Details
Select a transformation of the correlation coefficient	specifies the underlying distribution assumed for the test statistic. You can choose from a Fisher’s z transformation or a t transformation. If you select a t transformation, you must specify whether to use fixed variables or random (bivariate normal) variables.
Select the number of sides of the test	specifies the number of sides (or tails) and the direction of the statistical test. Here are the valid tests: a one-sided test, with the alternative hypothesis in the same direction as the effect. an upper one-sided test. The alternative hypothesis indicates a correlation greater than the null value. a lower one-sided test. The alternative hypothesis indicates a correlation less than the null value. a two-sided test
Significance Level
specifies the alpha value of the statistical test. The default is 0.05. Click to specify multiple values. Note: Specify only one value per row.
Correlation
Select a form	specifies whether to calculate the correlation or the correlation and the null correlation. Correlation specifies the correlation between two variables, possibly adjusting for any partial variables. Null correlation specifies the null value of the correlation. The default value is 0. Note: This option is available only if you select a Fisher’s z transformation.
Correlation values	specifies the correlation values. Click to specify multiple values. Note: Specify only one value per row.
Null correlation values	specifies the value of the null correlation. Note: This option is available only if you select a Fisher’s z transformation and the Correlation, null correlation option from the Select a form drop-down list.
Sample Size
Allow fractional sample sizes	enables fractional input and output for sample sizes.
Sample size values	specifies the sample size to use in the analysis.
Partial Variables
Specify the number of variables to partial out	specifies the number of variables adjusted for the correlation between the two primary variables. The default value is 0, which corresponds to a simple correlation.