Multiple Correspondence Analysis, which may be
considered an extension of the simple Correspondence Analysis, allows us to analyse the
relationships between two or more categorical variables.
In T-LAB, the limitations of this kind of
analysis are the following:
- 150,000 elementary contexts as rows;
- 250 variable categories as columns;
- 3,000 key-words, as supplementary columns (Lebart L., Salem A.,
Multiple Correspondence Analysis, available only if the
corpus includes at least two variables, requires that the user
select his options within the following window:
At the end of the analysis:
- T-LAB outputs are the same as correspondence
analysis (see below) plus the Burt table (Burt_Table.xls) including
all crossed variables;
- only when the elementary contexts
correspond to primary documents (e.g.
responses to open-ended questions) it is possible to do a cluster analysis.