In SEM, what is the physical significance of eigen value while doing principal component analysis (PCA)?
Hello Parag Sen , In SEM, you confirm the factor structure and not explore. PCA/EFA aren't part of SEM. CFA is a part of SEM.
% of common-factor variance explained by target factor.
Hope this helps,
Matt
The PCA is based on the eigenvalues of the variance-covariance matrix or the autocorrelation matrix of the data. The sum of all eigenvalues is equal to the overall variance. Therefore the eigenvalue of a variable is expressed as a fraction of the overall variance. Arranging the eigenvalues in order of magnitude the cumulative fractions are determined and used as basis for stoppage of the process.
The eigenvalues provides information regarding possible components/factors and their relative explanatory power. Also, eigenvalues assist in selecting the number of components/factors to retain for subsequent analysis. That is if the researcher wants to base on latent root criterion of retaining factors with eigenvalues greater than 1.0.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling