The DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index.
A Review of Exploratory Factor Analysis Decisions and Overview of Current Practices: What We Are Doing and How Can We Improve?
Abstract
Authors within the fields of cyberpsychology and human−computer interaction have demonstrated a particular interest in measurement and scale creation, and exploratory factor analysis (EFA) is an extremely important statistical method for these areas of research. Unfortunately, EFA requires several statistical and methodological decisions to which the best choices are often unclear. The current article reviews five primary decisions and provides direct suggestions for best practices. These decisions are (a) the data inspection techniques, (b) the factor analytic method, (c) the factor retention method, (d) the factor rotation method, and (e) the factor loading cutoff. Then the article reviews authors’ choices for these five EFA decisions in every relevant article within seven cyberpsychology and/or human–computer interaction journals. The results demonstrate that authors do not employ the recommended best practices for most decisions. Particularly, most authors do not inspect their data for violations of assumptions, apply inappropriate factor analytic methods, utilize outdated factor retention methods, and omit the justification for their factor rotation methods. Further, many authors omit altogether their EFA decisions. To rectify these concerns, the current article provides a step-by-step guide and checklist that authors can reference to ensure the use of recommended best practices. Together, the current article identifies concerns with current research and provides direct solutions to these concerns.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Hence, items (9, 17) were deleted.”
From Attitudes of faculty members in Palestinian universities toward employing artificial intelligence applications in higher education: opportunities and challenges (2024) · cites this paper as Howard (2016)
In the citing paper
“Exploratory factor analysis (EFA) revealed two dimensions, with sufficient indicator loadings onto the respective dimensions (Howard, 2016).”
From College students’ literacy, ChatGPT activities, educational outcomes, and trust from a digital divide perspective (2026) · cites this paper as Howard, 2016
In the citing paper
“The most common methods for factor analytics are principal components analysis (PCA), principal axis factoring (PAF), and maximum likelihood (ML) [48].”
From Data Science and Analytics: An Overview from Data-Driven Smart Computing, Decision-Making and Applications Perspective (2021) · cites this paper as [48]
In the citing paper
“The data met standard assumptions for PLS-SEM analysis [56]: significant Bartlett's test of sphericity on all constructs ( 𝜒 2 (496)=4474.58, p < .001) and adequate KMO measure of sampling adequacy (0.9), exceeding the 0.6 threshold [69].”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [69]
In the citing paper
“EFA identifies the optimal number of factors and their item groupings without imposing a predetermined structure [69].”
From What Needs Attention? Prioritizing Drivers of Developers' Trust and Adoption of Generative AI (2025) · cites this paper as [69]
Citations
Cites
Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing
Benjamini, Y., & Hochberg, Y. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02031.x · Cited by 109,482 · cited 1×
“If an author makes these decisions during the scale creation or evaluation process, such as through testing multiple methods, they may inadvertently increase their familywise error rate and risk committing a Type I error (Benjamini & Hochberg, 1995).”
MethodTabachnick, B. G., & Fidell, L. S. (1983) · Cited by 80,031
The application of electronic computers to factor analysis
Kaiser, H. F. (1960) · Educational and Psychological Measurement · doi:10.1177/001316446002000116 · Cited by 12,459 · cited 1×
“Possibly the first factor retention method is the Kaiser (1960) criterion, which suggests that all factors with eigenvalues above 1 should be retained.”
MethodCostello, A. B., & Osborne, J. (2005) · University of Massachusetts Amherst · doi:10.7275/jyj1-4868 · Cited by 10,297 · cited 10×
“…sample size are 200 to 500 participants (depending on communalities and other factors; Comrey & Lee, 1992; MacCallum, Widaman, Zhang, & Hong, 1999) and between a 5-to-20 and participant-to-variable ratio (Costello & Osborne, 2005; Hair, Black, Babin, Anderson, & Tatham, 2006; Velicer & Fava, 1998).”
BackgroundMethodInfluential(2007) · Cited by 10,092
Evaluating the use of exploratory factor analysis in psychological research.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999) · Psychological Methods · doi:10.1037/1082-989x.4.3.272 · Cited by 9,587 · cited 6×
“Further yet, authors have consistently condemned the use of PCA for several reasons (Fabrigar et al., 1999; Preacher & MacCallum, 2003).”
BackgroundMethodInfluentialA Rationale and Test for the Number of Factors in Factor Analysis
Horn, J. L. (1965) · Psychometrika · doi:10.1007/bf02289447 · Cited by 8,714 · cited 1×
“Instead, they represent statistical artifacts (Horn, 1965; O’Connor, 2000).”
BackgroundThe Varimax Criterion for Analytic Rotation in Factor Analysis
Kaiser, H. F. (1958) · Psychometrika · doi:10.1007/bf02289233 · Cited by 7,580 · cited 3×
“For this reason, Kaiser (1958) created varimax rotation.”
MethodInfluentialSample size in factor analysis.
MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999) · Psychological Methods · doi:10.1037/1082-989x.4.1.84 · Cited by 5,497 · cited 1×
“…most popular recommendations for minimum sample size are 200 to 500 participants (depending on communalities and other factors; Comrey & Lee, 1992; MacCallum, Widaman, Zhang, & Hong, 1999) and between a 5-to-20 and participant-to-variable ratio (Costello & Osborne, 2005; Hair, Black, Babin,…”
BackgroundA Second Generation Little Jiffy
Kaiser, H. F. (1970) · Psychometrika · doi:10.1007/bf02291817 · Cited by 4,785 · cited 4×
“The two most popular data inspection techniques for EFA are Bartlett’s test of sphericity (Bartlett, 1950; Dziuban & Shirkey, 1974) and the Kaiser–Meyer–Olkin (KMO) Measure of Sampling Adequacy (Dziuban & Shirkey, 1974; Kaiser, 1970).”
BackgroundMethodInfluential
Research works printed but not linked: 6
- 10.1037/0033-2909.103.2.276 · doi:10.1037/0033-2909.103.2.276 · Paper page
- Costello A. B., 2005, Practical Assessment, Research & Evaluation, 10, 1
- Fabrigar L. R., 2012, Understanding statistics: Exploratory factor analysis
- Hair J. F., 2006, Multivariate data analysis (Vol. 6)
- 10.1037// 1082–989X.4.1.84 · doi:10.1037//
- Tabachnick B. G., 2007, Using multivariate statistics
Cited by
Assessing measurement model quality in PLS-SEM using confirmatory composite analysis
Hair, J. F., Howard, M. C., & Nitzl, C. (2020) · Journal of Business Research · doi:10.1016/j.jbusres.2019.11.069 · Cited by 5,446
“Based on our review it is most likely neither side is entirely correct and both approaches are appropriate for specific research applications (Brown, 2014; Grimm & Yarnold, 1995; Hair, Black, et al., 2019; Howard, 2016; Kline, 2014; Kline, 2015; Thompson, 2004).”
BackgroundMethodSelf-citationVariable-Centered, Person-Centered, and Person-Specific Approaches: Where Theory Meets the Method
Howard, M. C., & Hoffman, M. E. (2018) · Organizational Research Methods · doi:10.1177/1094428117744021 · Cited by 1,121
“The p-EFA did not identify certain subpopulations within the sample or separate participants based on the probability of their belonging to these subpopulations.”
MethodInfluentialSelf-citationSarker, I. H. (2021) · SN Computer Science · doi:10.1007/s42979-021-00765-8 · Cited by 570
Law, E. L.-C., Brühlmann, F., & Mekler, E. D. (2018) · Proceedings of the 2018 Annual Symposium on Computer-Human Interaction in Play · doi:10.1145/3242671.3242683 · Cited by 196
“20 rule [23] states that an acceptable item should load at least .”
BackgroundMethodHoward, M. C., Boudreaux, M., & Oglesby, M. (2024) · European Journal of Work and Organizational Psychology · doi:10.1080/1359432x.2024.2393462 · Cited by 178
“When performing an EFA, factors reproduce the common variance of indicators (Howard, 2016; Howard & Henderson, 2023).”
Self-citation
Metadata
- Authors
- Howard, M. C.
- Affiliations
- Pennsylvania State University (US)
- Corresponding author
- Howard, Matt C.
- ORCID
- Howard, Matt C.
- Venue
- International Journal of Human-Computer Interaction
- Publisher
- Informa UK Limited
- Published
- 2015-09-14
- Volume
- 32
- Issue
- 1
- Pages
- 51-62
- ISSN
- 1044-7318 (ISSN-L)
- 1532-7590
- OpenAlex
- W2222174245
- Cited by
- 1,349
- References
- 60
- Citations by year
- 2026: 265 · 2025: 250 · 2024: 218 · 2023: 214 · 2022: 138 · 2021: 125 · 2020: 68 · 2019: 37 · 2018: 26 · 2017: 7 · 2016: 1
- FWCI
- 14.15
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- International Journal of Human-Computer Interaction (publishedVersion)
- Topics
- Social Sciences › Decision Sciences › Management Science and Operations Research › Psychometric Methodologies and Testing
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Psychology Research and Bibliometrics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Keywords
- exploratory factor analysis, factor retention, factor rotation, data inspection, scale development, cyberpsychology, human-computer interaction, best practices
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A Review of Exploratory Factor Analysis Decisions and Overview of Current Practices: What We Are Doing and How Can We Improve? https://app.cochat.ai/papers/10.1080/10447318.2015.1087664
- The paper
- Howard, M. C. (2016). A Review of Exploratory Factor Analysis Decisions and Overview of Current Practices: What We Are Doing and How Can We Improve? International Journal of Human-Computer Interaction. https://doi.org/10.1080/10447318.2015.1087664
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.