I have problems understanding the Factor Component Analysis of the paper: "Measuring thirty facets of the Five Factor Model with a 120-item public domain inventory: Development of the IPIP-NEO-120".
The paper says on page 86:
The two large Internet samples were used to examine the factor structure of the IPIP-NEO-300 and IPIP-NEO-120. Three principal components analyses were conducted. In each case, five components were extracted and subjected to varimax rotation with Kaiser normalization. The first analysis entered scores from the 30 10-item facet scales, the second used 4-item facet scale scores computed from the IPIP-NEO-300, and the third used 4-item facet scale scores from the IPIP-NEO-120. Results are displayed in Table 4. Although the usual five major personality factors are clearly represented by the loadings in all three analyses, not all facets show their highest loading on the expected component. For example, in all three analyses, N4 Self-Consciousness shows a stronger (negative) loading on the Extraversion factor than the Neuroticism factor. The difference in loadings was less pronounced for the analysis of the 10-item facet scales (.54 and .51 for E and N, respectively) than for the 4-item scales scored from the IPIP-NEO-300 (.71 and .23) and the 4-item scales from the IPIP-NEO-120 (.73 and .00).
The example from the text with N4 Self-Consciousness has in all three PCAs the highest absolute number at 1.
How does one know that number 1 corresponds to the Extraversion? They say five components were extracted and subjected to varimax rotation with Kaiser normalization. No clue what this means. In my opinion 1 would correspond to Neuroticism, 2 to Extraversion and so on.