Canonical Variates Analysis Options
See Also
This menu selects the information to be printed from a canonical variates analysis.

Display

This specifies which items of output are to be produced by the analysis.
Latent Rootslatent roots (eigenvalues)
LoadingsCanonical Variate Loadings (latent vectors/eigenvectors)
ResidualsResiduals (see note below)
Significance TestsSignificance Tests for equality of latent roots
Canonical Variate MeansCanonical variate means and between-group Mahalanobis distances

Number of Dimensions

Specifies how many latent roots and vectors are printed (and saved); the residuals are formed from the remaining dimensions.

Graphics

Specifies which graphical outputs are to be produced by the analysis.

Canonical Variate Plot
Draws a canonical variate plot of the scores given by the X Dimension and Y Dimension boxes. The display for the plot can be controlled using the following options:

X DimensionThe dimension to be plotted on the x-axis. The maximum dimension that can be displayed is displayed in the Number of Dimensions option.
Y DimensionThe dimension to be plotted on the y-axis. The maximum dimension that can be displayed is displayed in the Number of Dimensions option.
Display Confidence RegionWhether to include a confidence region on the plot. Selecting mean will draw the confidence region about the population means and the setting population plots the tolerance region for the populations.
Confidence limit (%)Allows you to control the confidence limit for the confidence region. The limit is expressed as a percentage and must be in the range 0-100

Action Buttons

OKSave the option settings and close the dialog.
CancelClose the dialog without making any changes.
DefaultsReset the options to their default settings.

See Also