Correspondence Analysis Results - Review Tab

Select the Review tab of the Correspondence Analysis Results dialog box to access the options described here. Use these options to review the input table (click the Observed frequencies button) as well as various values computed based on the input table. Note that you can mark these results spreadsheets as input spreadsheets for subsequent analysis, for example, to use the Graphics options to produce graphical summaries of the results.

Note: The default Graphs of Input Data for all tables is the 3D Histogram of the respective table entries.
Option Description
Observed frequencies Displays a spreadsheet showing the input table (example, frequencies in the two-way table).
Row percentages Displays a spreadsheet showing the values in the input table after transformation to row percentages. The values in each row sum to 100%. Note that the coordinates for the row points are computed based on this table if you select Row & column profiles or Row profiles (interpret row dist.) from the Standardization of coordinates group box on the Options tab as the standardization method. See Correspondence Analysis Results dialog box - Options tab for more details on standardization methods.
Column percentages Displays a spreadsheet showing the values in the input table after transformation to column percentages. The values in each column sums to 100%. Note that the coordinates for the column points are computed based on this table if you select Row & column profiles or Row profiles (interpret row dist.) from the Standardization of coordinates group box on the Options tab as the standardization method. See Correspondence Analysis Results dialog box - Options tab for more details on standardization methods.
Total percentages Displays a spreadsheet showing the values in the input table after transformation to percentages of the total sum of all values. The sum of all values in the table equals to 100%.
Expected (chi-square) Displays a spreadsheet with the expected values under the hypothesis that the row and column variables are independent of each other. The standard formula is used for computing the expected values, based on the marginal totals of the input frequency table. For example, for row i and column j the expected value is computed as the row total for row i, times the column total for column j, divided by the grand total.
Observed minus expected Displays a spreadsheet with the differences between the observed table values and the expected values.
Contributions to chi-square Displays a spreadsheet with the respective cells' contributions to the overall Chi-square value (note that these values are only valid if the input table contains frequencies). For example, for row i and column j this value is computed as: (observed value - expected value)² /expected value. The sum of all values is equal to the overall Chi-square value.
Standardized deviates Displays a spreadsheet whose standardized deviate values are computed as the square-root of the contribution to Chi-square values.