Factor analysis is a statistical technique used to explain the variance observed in a set of observed variables in terms of fewer unobserved variables called factors. The assumption is that there are underlying latent factors that influence multiple measured variables, causing them to correlate. For example, in psychology, responses to many test questions might be driven by a few underlying traits like “extroversion” or “intelligence”. Factor analysis finds factors by looking for sets of variables that move together (are highly correlated), often using methods like principal axis factoring or maximum likelihood. The output includes factor loadings (which variables are associated with which factors) and sometimes factor scores (estimates of factor values for each observation). It’s an unsupervised dimensionality reduction technique, similar to PCA but focused on modeling covariance structure rather than maximizing variance.