There are two general types of factor analysis: exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). The most important difference is that CFA has clear expectations on a specific factor structure, which is what we test, whereas EFA does not rely upon any expected structure. In this guide, we will focus on EFA (hereafter referred to simply as factor analysis). If you are interested in learning more about CFA, we suggest that you look up structural equation modelling (SEM), which is a very useful framework.

More information help sem

The main feature of factor analysis is that is enables us to investigate the underlying structure in the pattern of correlations between a number of variables (often referred to as “items”). There are many different ways of using factor analysis, but one very practical application is cases where we have several items from a questionnaire that we want to create an index for. By conducting a factor analysis, we are able to see whether the items represent the same factor (or “dimension”). If so, we can create our index. Factor analysis can also tell us how to improve our index (e.g. by excluding one or more items), or if we actually have more than one factor and thus need to consider creating separate indices.