Exploratory Factor Analysis

By Joseph C. Cappelleri, Robert A. Gerber. Applied Univariate, Bivariate, and Multivariate Statistics Using Python, 2018

1,904 citations recorded by the source corpus.

PCA and SVD are considered simple forms of exploratory factor analysis. The term ‘factor analysis’ is a bit confusing and you will find a variety of definitions out there–some people assert that PCA is not factor analysis, and others might use PCA but call it factor analysis. PCA involves a complete redescription of the covariance or correlation matrix along a set of independent dimensions, until it completely redescribes the data, and then you pick off the most important components and ignore the rest. EFA is typically attempting to do the same thing, but trying to maximize the variance accounted for by a fixed number of ‘latent’ factors, so that we have the model + error. Furthermore, most of the time factor analysis lets us pick orthogonal dimensions based on criteria other than maximum variance, and permits factors that are not orthogonal, allowing for latent factors that are somewhat related.

Subfields: Food Science, Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/21877990650a820ac946f5fd2a2cf8085114ae73

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Food Science

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

A review of the methodological misconceptions and guidelines related to the application of structural equation modeling: A Malaysian scenario (2017, neutral): “We would recommend some good articles on the subject matter, and they are authored by Hair et al. (2017a), Fan et al. (2016), Green et al. (2016), Brown (2015), Fabrigar and Wegener (2011) and Winter, Dodou, and Wieringa (2009).”
One Size Doesn’t Fit All: Using Factor Analysis to Gather Validity Evidence When Using Surveys in Your Research (2019, positive): “FACTOR ANALYSIS: EVIDENCE OF DIMENSIONALITY AMONG A SET OF ITEMS Factor analysis is a statistical technique that analyzes the relationships between a set of survey items to determine whether the participant’s responses on different subsets of items relate more closely to one another than to other su”
Reliability and Validity Evidence for the Institutional Integration Scale (2004, background): “Deletion due to low–factor pattern coefficients was not sufficient evidence to eliminate the items (Goodwin & Goodwin, 1999), especially because factoring individual items can be problematic (Gorsuch, 1988).”
Genomic relationships, novel loci, and pleiotropic mechanisms across eight psychiatric disorders (2019, positive): “A follow-up conﬁrmatory model with three correlated factors was speciﬁed in Genomic SEM based on the EFA parameter estimates (positive standardized loadings > 0.2 were retained; Figure 2B).”
Mathematics Anxiety: What Have We Learned in 60 Years? (2016, neutral): “Good reliability has also been found for other mathematics anxiety measures such as Betz’s (1978) Mathematics Anxiety Scale (Dew et al., 1984; Pajares and Urban, 1996) and the Fennema–Sherman scales (Mulhern and Rae, 1998).”
A journey around alpha and omega to estimate internal consistency reliability (2017, neutral): “Raykov and Marcoulides (2015) where, as usual, these authors include the Mplus syntax to calculate the reliability of the scale scores and their CI.”
Exploratory Factor Analysis: A Guide to Best Practice (2018, positive): “Although not universally agreed (Gorsuch, 1983), pattern coefficients should be the first focus of interpretation in most analyses (Cudeck, 2000; Fabrigar & Wegener, 2012; Hair et al., 2010; Price, 2017).”
Investigating the performance of Exploratory Graph analysis and traditional techniques to identify the number of latent factors: a simulation and tutorial (2018, methodology): “The PC criterion varies from 0% (signaling complete inaccuracy) to 100% (indicating perfect accuracy). In the case of the MBE, 0 reflects a total lack of bias, while negative and positive values denote underfactoring and overfactoring, respectively. Regarding the MAE criterion, higher values signal ”
A Review of Exploratory Factor Analysis Decisions and Overview of Current Practices: What We Are Doing and How Can We Improve? (2016, neutral): “Once again, these mathematics are not described within the current article, but they have been detailed elsewhere (Cudeck, 2000; Gorsuch, 1988).”
Applied Psychometrics: The Steps of Scale Development and Standardization Process (2018, neutral): “In most cases a common practice in test development involves a combination either of confirmatory factor analysis (CFA) and IRT (Irwing & Hughes, 2018) or more commonly a combination of EFA and CFA (Steger et al., 2006; Fabrigar & Wegener, 2012; Kyriazos, 2018a).”
MAQuA: Adaptive Question-Asking for Multidimensional Mental Health Screening using Item Response Theory (2025, unknown): “All participants then took ten rating scale questionnaires along with language-based questions related to all the mental health conditions considered.”