Product-Moment and Partial Correlations - Options tab

Select the Options tab of the Product-Moment and Partial Correlations dialog box to access options to affect the display of correlations as well as the calculations.

Display format for correlation matrices. These options affect the display format of both correlation matrices and partial correlation matrices.

Display simple matrix (highlight p's). When this option button is selected, clicking the Summary button produces a matrix of correlation coefficients with the significant correlations highlighted in the spreadsheet. Use the p-value for highlighting option (described below) to change the default p (alpha) value (default p = .05).

Display r, p-values, and N's. When this option button is selected, clicking the Summary button produces an augmented summary correlation matrix; that is, each cell in the matrix does not only contain the respective correlation coefficient but also the p-value, r and valid N (if the Casewise deletion of missing data option button is not selected). Again, significant correlations (cells in the matrix) are highlighted in the spreadsheet and you can use the p-value for highlighting option (described below) to change the default p (alpha) value.

Display detailed table of results. When this option button is selected, clicking the Summary button produces a detailed table of results (not in matrix format). This table reports not only the correlation coefficients but also the pairwise means, pairwise standard deviations, regression weights, intercepts, and other regression statistics.

Display long variable names. Select this check box to display the long variable names (if any, see Variable Specs Editor) along with the short names in the first column of the results spreadsheets. If no long variable names have been specified for any of the selected variables, the setting of this check box has no effect.

Extended precision calculations. Select this check box to use extended precision calculations in your descriptive statistics. Statistica employs proprietary extended precision algorithms in order to minimize round-off error in the computations. This option should be selected if the variables in the analysis show extremely small relative variances (variance divided by the mean), for example, if all values for a variable range between 10000.000001 and 10000.000002 (note that there is a difference between small variance and small relative variance; e.g., the relative variance of values in the range of .0000000000000001 to .0000000000000002 is not small). For extremely large data files with many thousands of cases, the computations may be somewhat slower when this check box is set.

p-value for highlighting. The default p-value for highlighting in correlation matrices is .05. You can adjust this p-value by entering a new value in the edit box or using the microscroll buttons. For more details on p-value, see Elementary Concepts.

Include means and std. devs. in square matrices. Select this check box to include the means and standard deviations of all variables in the results spreadsheets. When this check box is selected, the mean and standard deviation for each variable is calculated and appended to the end of the correlations spreadsheet. Additionally, correlations are reported with a greater degree of accuracy (i.e., 9-digit precision versus 2-digit precision when means and standard deviations are not reported). Note that this option is only applicable for square (or one variable list) matrices. Note also that you can increase (or decrease) the display of decimals for a selection of variables by clicking the Increase Decimal (or Decrease Decimal) toolbar button. You can also control the display format for a single variable using options in the variable specification dialog box.