Factor Analysis

Abstract

Factor analysis is a term used to refer to a set of statistical procedures designed to determine the number of distinct unobservable constructs needed to account for the pattern of correlations among a set of measures. These unobservable constructs that explain the pattern of correlations among measures are referred to as common factors. The statistical procedures comprising factor analysis provide information regarding the number of common factors underlying a set of measures as well estimates of the strength and direction of influence of each of the factors on each of the measures. These estimates of influence are referred to as factor loadings. In cases where there is a lack of clear expectations regarding the number and nature of the factors likely to underlie a set of measures, procedures exist to conduct exploratory factor analysis (EFA) or unrestricted factor analysis. In this bibliography, we will primarily focus on these exploratory procedures. In cases in which a researcher can make clear predictions regarding the number of factors and the specific measures each factor will influence, procedures are available to conduct confirmatory factor analysis (CFA) or restricted factor analysis. Confirmatory factor analysis is typically covered in reviews of structural equation modeling, and thus we will draw upon references from CFA only when they have direct relevance to EFA.