Statistics by Groups Results - Correlations tab

Select the Correlations tab in the Statistics by Groups Results dialog box to access options to review and compare the correlations.

Within-group correlations & covariances

If more than one dependent variable is selected for the current analysis (via the Variables button), click the Within-group correlations & covariances button to compute the within-group correlation matrices for those variables.

The Select groups dialog box is displayed in which you select the group for which to display the correlation matrix, or you can select to display a cascade of spreadsheets (displaying the correlation matrices for multiple groups).

The default results spreadsheet graph (via Graphs of Input Data) for this option is the scatterplot. From within the spreadsheet containing a within-group correlation matrix, you can produce a scatterplot of selected within-group correlations.

Options

This group box contains options for the Within-group correlations matrices (see above).

Specifically, you can determine an alpha (p) value for highlighting significant correlations, request to compute within-group covariance matrices, or report various additional statistics (such as pairwise N, means, standard deviations).

Compute & display significance (p)-values

Select this check box to compute a matrix of p significance values for each correlation matrix.

Compute & display covariances

Select this check box to compute a matrix of covariances for each correlation matrix.

Options for pairwise MD deletion

The options in this group box are meaningful and available only if pairwise deletion of missing data is selected prior to processing the data.

Display pairwise N

Select this check box to display a matrix of valid N for each correlation matrix.

Display pairwise means

Select this check box to display a matrix of pairwise means for each correlation matrix.

For each correlation, those means are computed from all cases that are included in the computation of the respective correlation coefficient.

Display pairwise standard deviations

Select this check box to display a matrix of pairwise standard deviations for each correlation matrix.

For each correlation, those standard deviations are computed from all cases that are included in the computation of the respective correlation coefficient.

p-value for highlighting

The default p-value for highlighting is .05.

You can adjust this p-value by entering a new value in the edit box or using the microscrolls. For more details on p-value, see Elementary Concepts.

Categ. scatterplots

Click this button to produce a cascade of categorized scatterplots.

A standard variable selection dialog box is displayed, in which you select variables for the plot from among the list of dependent variables in the current analysis. A categorized scatterplot is produced for each selected variable.

Save stacked matrix

Click this button to display the Select groups dialog box, which is used to select the groups you want to include in your stacked matrix (also called a multiple correlation matrix).

After you have selected the groups to be displayed, the stacked matrix spreadsheet is produced in an individual window (regardless of the settings on the Options dialog box - Output Manager tab or the Analysis/Graph Output Manager dialog box). You can, however, add this spreadsheet to a workbook or report using the Add to Workbook or Add to Report buttons on the Home tab. Note that in order to save the stacked matrix spreadsheet, you must select the spreadsheet and select Save or Save As on the File tab. This is useful if you want to use the stacked matrix for further analyses with other Statistica analyses.

See also: Breakdown: Descriptive Statistics by Groups