Introduction to Factor Analysis

Abstract

Factor analysis is a multivariate statistical method wherein we strive to uncover a structure or patterns in the associations between variables (e.g., items of a questionnaire) and represent them more parsimoniously with a smaller set of underlying latent variables, called factors. These factors are thought to constitute unobservable, internal attributes, that influence or cause the way observable, i.e., mani-fest behavior is expressed. For example, intelligence is thought to constitute such an attribute: If a person’s intelligence is high, this should allow them to response correctly to hard questions in intelligence tests and be fast at learning about complex new topics.

Factor analysis is used to describe and test (hypothesized) structures and patterns in data. For in-stance, in personality psychology, we might look at data from a big five questionnaire and test whether we actually do find five distinct factors, as proposed by the theory; or in an instrument assessing ADHD, we can test whether we can identify distinct factors for the main symptoms and whether they match with the empirically distinguished subtypes; or, finally, in a questionnaire assessing risk preferences across different life domains, we can test whether we find a stronger clustering of items within vs. across domains. Factor analysis is also the basis for structural equation modelling, wherein we can run regressions (i.e., test directed associations) between latent constructs instead of running a regression based on manifest variables.

In this workshop we cover the basics of exploratory and confirmatory factor analysis including their application in R. Moreover, I provide a sneak peek into structural equation modelling. My goal is that after the workshop you will know which technique to employ in a given scenario; that you can take the first steps of implementing it in R; and that you know where to look for further resources and details to be able to successfully apply factor analysis in your research. To this end, I will include many examples and exercises alongside the theoretical introduction.