Factor Analysis

Peter Schmidt & Johannes Herrmann

Factor analysis is a well-established method that attempts to measure latent constructs such as attitudes or values, which cannot be observed directly. The standard model assumes that the measured variables (items or indicators) are linear additive functions of the unobserved (latent) factors and the error component. This is called the common-factor model. The general system of equations for this model is where y j represents the 1, …, j observed variables measured on a sample of n independent subjects; η 1 , …,η m represent the 1, …, m latent constructs (factors) in the model; λ j 1 , …, λ m represent the factor loading (partialized regression coefficient) relating variables 1, …, j to the first to m th factors; and ε 1 , …, ε j stands for the error component (uniqueness). It is assumed that the error component of one indicator or item is independent of An ...

Users without subscription are not able to see the full content on this title. Please, subscribe or login to access all content on this website.