Multivariate Analysis: Factor Analysis |
The Rotation Tab
You can use the Rotation tab (Figure 27.5) to
transform the factors by orthogonal or oblique rotations.
Orthogonal rotations rigidly rotate the factors; oblique transformations
introduce correlations between the factors. Transformed factors are
often more interpretable in terms of the original variables.
- Factor rotation
-
specifies the rotation method. You can select from a set of common
orthogonal or oblique transformations.
This corresponds to the ROTATE= option in the PROC FACTOR statement.
- Harris-Kaiser power
-
specifies the power of the square roots of the eigenvalues used to
rescale the eigenvectors for Harris-Kaiser orthoblique transformation.
This corresponds to the HKPOWER= option in the PROC FACTOR statement.
- Promax power
-
specifies the power for forming the target Procrustean matrix.
This corresponds to the POWER= option in the PROC FACTOR statement.
- Factor pattern normalization
-
specifies the method for normalizing the rows of the factor pattern
for rotation.
This corresponds to the NORM= option in the PROC FACTOR statement.
Copyright © 2008 by SAS Institute Inc., Cary, NC, USA. All rights reserved.