The PCA method that we have discussed so far models all of the variance of the variables to which it is applied. An alternative approach, which is often confused with PCA, is to model only the common variance: an approach called factor analysis (FA). In this chapter, we will discuss exploratory factor analysis (EFA).
The following are the basic terminologies that you need to be aware of:
Latent trait or common factor: This is an unobserved variable that explains some or all of the variance in observed variables.
Path coefficient: This is the correlation coefficient between a latent trait and an observed variable.
Communality: This refers to the square of a path coefficient in a single factor model.
Uniqueness: Computationally, this is simply one minus the communality of an observed variable. (If a covariance matrix is used, it is equal to variance minus one.)