Multiple Correspondence Analysis (MCA)

Complete implementation of Multiple Correspondence Analysis (MCA) for Burt tables; see also the separate option for simple Correspondence Analysis (CA). Various methods for specifying input data are supported, including the direct input of Burt tables.

General

Element Name Description
Detail of computed results reported Specifies the level of computed results reported. At the Minimal level of detail, Statistica will report the summary coordinates only. If Comprehensive results is requested, plots of coordinates are reported as well, along with the observed input table. If All results is selected, a large number of other diagnostic statistics are also reported (percentages, expected and residual values, etc.).
Input data format Select whether the input data describes raw data or frequency counts with grouping variables. If Raw data is selected, Statistica expects categorical variables with codes that identify to which group each observation belongs; Statistica will then compute the input table by tabulating the codes found in those variables. If Frequencies with grouping is requested, Statistica expects grouping variables as well as a (continuous) variable with frequency counts (if more than one is specified, then separate analyses are performed for each).
Method, number of dimensions Specifies whether to extract a fixed number of dimensions, or to use a cutoff value for the cumulative contribution to inertia in order to determine the number of dimensions for the final results.
Number of dimensions Specifies the number of dimensions for which to compute coordinates (for spreadsheets and plots of points); this option is only applicable if the Method for determining the number of dimensions was set to Number of dimensions.
Cumul. contribution to inertia Specifies the total cumulative contribution to inertia at which point no additional dimensions are to be extracted; this option is only applicable if the Method for determining the number of dimensions was set to Cumulative contribution to inertia.