Correspondence Analysis

when called separately, requires as input a file containing a matrix of occurrences of the vocabulary items (columns) by the context units identified for a particular text (columns), which is optionally generated by HAMLET Joint Frequencies. Alternatively, the profiles of occurrences of items of a given search list in a number of different texts can be compared directly by selecting the option CA in the fast-track procedure

Correspondence Analysis represents the row and column categories of the input matrix of context unit profiles as points in the same dimensionality. The input matrix is first normalized by dividing each row entry by the square root of the product of the corresponding row and column totals, their geometric mean. This removes differences in the marginal totals and expresses each cell as a proportion. If any of these products is zero, a warning will be displayed and the corresponding rows or columns will be ignored in subsequent analysis.

  • When a valid file name is found, the vocabulary items available are listed as follows:

  • Click on   to start the process, or  to return to the previous window.

    It is important to check that the eigenvalues remaining after ignoring the first one are in fact large enough to justify continuing with the analysis. Reference can be made to the chi-squared contributions of each dimension of "inertia" reported, and to the overall chi-squared value for the analysis.

    The canonical scores are automatically displayed in graphic form.