Principal Components and Classification Analysis Results - Cases Tab

Principal Components and Factor Analysis

Select the Cases tab of the Principal Components and Classification Analysis Results dialog box to access options to produce PCCA results for cases. Various plotting options are available to provide a visual aid for interpretation of the results and classification of cases.

Factor coordinates of cases

Click Factor coordinates of cases button to display a spreadsheet with factor coordinates of cases. The spreadsheet also contains:

(1) an additional column displaying the code for active and supplementary cases, if a Variable for active cases was selected through the Variables button on the Principal Components and Classification Analysis Startup Panel - Quick tab;

(2) an additional column (last), displaying the label defined by the grouping variable, if a Grouping variable was selected through the Variables button on the Principal Components and Classification Analysis Startup Panel - Quick tab.

Note: The spreadsheet displays the factor coordinates only for the complete (valid) cases, if the Casewise option button was selected in the MD deletion group box on the Principal Components and Classification Analysis Startup Panel. Alternatively, if Mean substitution was selected, the mean of the corresponding variable is used in place of missing data and the spreadsheet contains the factor coordinates for all cases. The Number of factors (see the Principal Components and Classification Analysis Results dialog box) selected for the current analysis determines the number of factors for which the coordinates are computed and displayed. Note that the estimates of means and standard deviations that are used in the computation of factor coordinates of cases are computed from all valid cases, both active and supplementary. Refer to the Introductory Overview for additional details.

Plot case factor coordinates, 2D

Click the Plot case factor coordinates, 2D button to produce a 2D scatterplot of the factor coordinates for the cases. The plot contains coordinates for the supplementary cases, provided a Variable for active cases was selected and a valid Code for active cases was chosen on the Principal Components and Classification Analysis Startup Panel - Quick tab. After you click this button, the Select Two Factors for the Plot dialog box is displayed by using which you can select factors for the factor axes. Subsequent to selecting this button, you can use one of the Options for plot of factor coord. (see below).

Options for plot of factor coord

The options in the Options for plot of factor coord. group box pertain to the plot of the factor coordinates of cases (see Plot factor coordinates, 2D above). By default, when the input data file contains case names, the Case names option button is selected and the cases are labeled in the scatterplot by their respective case names. If the input data file does not contain case names, the Case numbers option button is the default. If you select the No names/numbers option button, cases are displayed only as points in the scatterplot, without their names or numbers. The Grouping labels option button is used to label the cases, provided the grouping variable was selected for the analysis. This option is often useful to identify and label the clusters of cases. Finally, if you want to project only those cases onto the factor plane for which the sum of cosine squares for the selected pair of factors is greater than a certain quantity C (0<=C<=1), enter this value in the Plot cases with sum of cosine² >= edit field. Note that the Plot cases with sum of cosine² >= field works in conjunction with the Number of factors box on the Principal Components and Classification Analysis Results dialog box.

Case names

Select the Case names option button to display the names of cases for which the factor coordinates are plotted.

Case numbers

Select the Case numbers option button to display the case numbers, instead of the case names. You might prefer to display the case numbers in the plot of factor coordinates if the case names are too long or the number of cases is too large.

No names/numbers

Select the No names/numbers option button to plot the case factor coordinates just as points, without the case names or numbers.

Grouping labels

Select the Grouping labels option button to display the label for each case coordinate. Such labels are defined by the Grouping variable, which can be selected through the Variables button on the Principal Components Classification Analysis Startup Panel - Quick tab. This option is, therefore, only available if a grouping variable was selected.

Factor scores

Click the Factor scores button to display a spreadsheet with factor scores. The spreadsheet may also contain:

1. an additional column displaying the code for active and supplementary cases, if a Variable for active cases was selected through the Variables button on the Principal Components Classification Analysis Startup Panel - Quick tab;
2. an additional column (last), displaying the label defined by the grouping variable, if a Grouping variable was selected through the Variables button on the Principal Components and Classification Analysis Startup Panel - Quick tab.

Note: The spreadsheet displays the factor scores only for the complete (valid) cases if the Casewise option button is selected in the MD deletion group box in the Principal Components Classification Analysis Startup Panel. Alternatively, if Mean substitution was selected, the spreadsheet shows zeros for the missing data cases. The Number of factors (see the Principal Components and Classification Analysis Results dialog box) selected for the current analysis determine the number of factors for which the coordinates are computed and displayed. Note that the estimates of means and standard deviations that are used in the computation of factor scores are computed from all valid cases, both active and supplementary. Refer to the Introductory Overview for additional details.

Factor score coefficients

Click the Factor score coefficients button to display a spreadsheet with the factor score coefficients.

Contributions of cases

Click the Contributions of cases button to display a spreadsheet with contributions of active cases to the variances, based on the first factor, first two factors, and so on. The contributions shown in the last column of this spreadsheet are the contributions computed from the current Number of factors (see the Principal Components and Classification Analysis Results dialog box topic).

Cosine²

Click the Cosine² button to display a spreadsheet with the cosine squares of the angle formed by the vector radius of each point and the factor axes. The values in this spreadsheet represent the relative contribution of the factor axes to the eccentricity of each case (see Introductory Overview). This spreadsheet also supplies information as to whether a particular case is an active or supplementary case. Each case is also labeled, if you have selected a Grouping variable through the Variables button on the Principal Components Classification Analysis Startup Panel - Quick tab. Note that in case of Casewise deletion of missing values [specified through the MD deletion group box on the Principal Components Classification Analysis Startup Panel], the Cosine² values for the missing data are not available. Refer to the Introductory Overview for additional details.

Save case statistics

Select either Factor coordinates or Factor scores and click the Save case statistics button to display a standard variable selection dialog box. After you select the variable(s), a spreadsheet containing the specified variable(s) along with the factor scores or coordinates is displayed in an individual window (regardless of the settings in the Options dialog box - Output Manager tab or the Analysis/Graph Output Manager dialog box). You can, however, add the spreadsheet to a workbook or report using the or buttons, respectively. Note that, to save the spreadsheet, you must select the spreadsheet and select Save or Save As from the File menu. This is useful if you want to use the factor scores for further analyses with other Statistica analyses.