statistical index (factor analysis) project index factor analysis decision tree

English or languish - Probing the ramifications
of Hong Kong's language policy

FACTOR ANALYSIS
Variable Loadings on Two Factors - A Sample Solution
factor rotation (rotated, orthogonal extraction) | graph (orthogonal and oblique rotation)

A Comparison Between Unrotated and Rotated Factor Solutions
  Unrotated Orthogonal Soution Orthogonally Rotated Solution
  Factor 1 Factor 2 Factor 1 Factor 2
Variable 1 0.50 0.80 0.03 0.94
Variable 2 0.60 0.70 0.16 0.90
Variable 3 0.90 -0.25 0.95 0.24
Variable 4 0.80 -0.30 0.84 0.15
Variable 5 0.60 -0.50 0.76 -0.13

In the unrotated sample solution variables 1, 2, and 5 load similarly on both factors, thus making it difficult to distinguish between them. By rotating the factor solution this ambiguity between the factors is removed. In the absence of ambiguity it is much easier to interpret and name the two factors. Obviously, whatever variance variables 3, 4, and 5 share in common, it is distinctly different from that which is shared by variables 1 and 2. A graphical depiction demonstrates more clearly what is meant by orthogonal rotation.

top