This questions is about Principal Component Analysis (PCA).

Based on following result of estimation PCA by using "Princomp" and "Principal" functions in R programming, please determine the Number of Components and compare the summary result of these functions. Why their estimations of loading are different? Which one is correct?

> principal(cor(course) , nfactor=3, rotate="none") > summary(princomp(course, cor=T) , loading=T) Principal Components Analysis Importance of components: Call: principal(r = cor(course), nfactors = 3, rc Comp. 1 Comp. 2 Comp . 3 Comp . 4 Comp . 5 Standardized loadings (pattern matrix) based upor Standard deviation PC1 PC2 PC3 h2 u2 com 1.7868732 1.0421767 0.69386264 0. 40661947 0.27233637 x1 0.77 -0.60 0.14 0.96 0.0397 2.0 Proportion of Variance 0.6385832 0.2172265 0.09628907 0.03306788 0.01483342 x2 0.87 -0.38 0.23 0.96 0.0426 1.5 Cumulative Proportion 0.6385832 0.8558096 0.95209870 0.98516658 1.00000000 x3 0.80 -0.03 -0.60 0.99 0.0074 1.9 x4 0.85 0.43 0.02 0.91 0.0886 1.5 Loadings : x5 0.70 0.63 0.23 0.94 0.0612 2.2 Comp. 1 Comp.2 Comp. 3 Comp. 4 Comp. 5 x1 0.429 0.571 0.195 0.261 0.619 PC1 PC2 PC3 x2 0. 488 0.365 0.330 -0.210 -0.690 ss loadings 3.19 1.09 0.48 X3 0.445 -0.863 0.189 -0.141 Proportion Var 0.64 0.22 0.10 Cumulative Var 0.64 0.86 0.95 x4 0.476 -0.416 -0. 696 0.340 Proportion Explained 0.67 0.23 0.10 x5 0.391 -0.605 0.327 0.607 Cumulative Proportion 0.67 0.90 1.00