Specify the Dimensions of the Table

Click the Specify structure of table button on the Multiple Correspondence Analysis (MCA) tab of the Multiple Correspondence Analysis (MCA): Table Specifications Startup Panel to display the Specify the dimensions of the table dialog. Use this dialog to specify the factor names in the Burt table and the number of levels for each factor. It is assumed that you typed the frequencies (or other measure of correspondence, confusion, similarity, association, etc.) for the table directly into the data file. Note that the sum of the levels that you specify must be equal to the number of variables you have selected. When processing the data, STATISTICA will automatically check whether the respective data specify a valid Burt table.

No. of levels. Specify the number of levels for each factor in the No. of levels fields. Note that the total number of levels must be equal to the number of variables reported in the Number of variables field.

Factor Name. Enter names for the factors in the Factor Name fields. If you do not provide a name, STATISTICA will assign a factor name for you.

For example, suppose your data file contains the following valid Burt table (see MCA Introductory Overview for a discussion of Burt tables):

You would specify 3 factors: for the first factor enter 2 in the No. of levels field, and name it, for example, Survival (i.e., type Survival in the Factor Name field). Specify the remaining two factors as Age with 3 levels, and Location with 3 levels.

OK. Click the OK button to accept the specified entries and return to the Multiple Correspondence Analysis (MCA) tab.

Cancel. Click the Cancel button to return to the Multiple Correspondence Analysis (MCA) tab. Any entries made in the Specify the dimensions of the table dialog will be ignored.

Note: Supplementary columns (variables). Specify the structure for the full Burt table via the Specify the dimensions of the table dialog, including any supplementary columns; that is, you need to "partition" the entire variable list into levels of factors.