Exploratory factor analysis: A practical guide for psychological research

Exploratory Factor Analysis (EFA) is a multivariate statistical method that aims to provide a parsimonious and simple representation (i.e., explanation) of the associations between a set of observed variables (i.e., manifest variables, indicators) through the smallest number of hypothesized latent variables or factors (i.e., constructs, dimensions) (e.g., Gorsuch, 1983; Mulaik, 1990). Given a set of observed variables in relation to one another, the EFA method aims to identify coherent groupings of these variables that are relatively independent of each other. Each of these groupings, according to the theoretical model of EFA, is caused by a latent factor. Given that, EFA aims to determine the number, nature, and inter-relations of latent variables, i.e., common factors, that explain these patterns of correlations between a set of observed variables (see, for example, Tabachnick & Fidell, 2019). To conduct an EFA, it is necessary to make several key decisions (see, for example, Fabrigar et al., 1999; Watkins, 2018; Norris & Lecavalier, 2010). This chapter aims to present easy-to-use guidelines and provide an example that demonstrates their application to facilitate the completion of an EFA. Our aim is to help readers conducting EFA and critically reviewing papers using EFA.