The Use of Exploratory Factor Analysis and Principal Components Analysis in Communication Research

Abstract

Exploratory factor analysis is a popular statistical technique used in communication research.Although exploratory factor analysis (EFA) and principal components analysis (PCA) are different techniques, PCA is often employed incorrectly to reveal latent constructs (i.e., factors) of observed variables, which is the purpose of EFA. PCA is more appropriate for reducing measured variables into a smaller set of variables (i.e., components) by keeping as much variance as possible out of the total variance in the measured variables. Furthermore, the popular use of varimax rotation raises some concerns about the relationships among the factors that researchers claim to discover. This paper discusses the distinct purposes of PCA and EFA, using two data sets as examples to highlight the differences in results between these procedures, and also reviews the use of each technique in three major communication journals: Communication Monographs, Human Communication Research, and Communication Research. Hee Sun Park (M.A., University of Hawai'i, 1998) is a visiting instructor in the Department of Communication at Michigan State University. René Dailey (M.A., University of Wyoming, 1998) and Daisy Lemus (B.A., University of Southern California, 2000) are graduate students in the

Verbal Fluency: Language or Executive Function Measure?

“…While some researchers may argue that principal components analysis (PCA) is a reasonable substitute for analyses of common factors (Velicer & Jackson, 1990), PCA was not applicable in this study because the objective of PCA is to reduce the measured variables to a smaller set of composite components (Park, Dailey, & Lemus, 2002). In addition, the use of EFA focuses on the shared variance among the variables by separating common variance from unique variance (Park et al, 2002), whereas PCA does not differentiate between common and unique variance. Factor extraction was determined using Horn's parallel analysis (Horn, 1965).…”

Moving beyond cognitive elements of ICT literacy: First evidence on the structure of ICT engagement

“…Factors resulting from EFA were extracted by considering the residuals, the amount of variance explained and the interpretability of the results (Asparouhov & Muthén, 2009;Park, Dailey, & Lemus, 2002). For the CFA, robust weighted least squares estimator was applied (WLSMV) using the diagonal weight matrix with robust standard errors and mean-variance-adjusted v 2 test statistics (Brown, 2006;Flora & Patrick, 2004).…”

Mind the Gap

“…Since our population is a relatively more homogeneous group than that studied by Weaver and colleagues (Weaver andWilhoit 1986, 1996;Weaver et al 2007), it is possible that members of this group define roles differently from other journalists not covering the political beat. We ran an exploratory factor analysis (EFA), which is the appropriate analysis to reveal latent constructs from observed variables (Park, Dailey, and Lemus 2002). Since we assume that the items measuring role conceptions can be correlated, we also used oblique (Promax) rotation with Kaiser normalization (Park, Dailey, and Lemus 2002;Field 2009).…”

“…We ran an exploratory factor analysis (EFA), which is the appropriate analysis to reveal latent constructs from observed variables (Park, Dailey, and Lemus 2002). Since we assume that the items measuring role conceptions can be correlated, we also used oblique (Promax) rotation with Kaiser normalization (Park, Dailey, and Lemus 2002;Field 2009). The analysis revealed four distinct factors that deviate from what Weaver and colleagues found (Weaver andWilhoit 1986, 1996;Weaver et al 2007).…”