A Brief Guide to Structural Equation Modeling
Abstract
To complement recent articles in this journal on structural equation modeling (SEM) practice and principles by Martens and by Quintana and Maxwell, respectively, the authors offer a consumer’s guide to SEM. Using an example derived from theory and research on vocational psychology, the authors outline six steps in SEM: model specification, identification, data preparation and screening, estimation, evaluation of fit, and modification. In addition, the authors summarize the debates surrounding some aspects of SEM (e.g., acceptable sample size, fit indices), with recommendations for application. They also discuss the need for considering and testing alternative models and present an example, with details on determining whether alternative models result in a significant improvement in fit to the observed data.
Integrity signals
Registry
Cites
Statistics
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?
A total of 489 valid samples were collected, which exceeded the minimum sample size recommended by Weston and Gore (2006).
“Finally, a total of 489 valid samples were collected, which exceeded the minimum sample size recommended by Weston and Gore (2006).”
From A study on the influence of rural tourism’s perceived destination restorative qualities on loyalty based on SOR model (2025) · cites this paper as Weston and Gore (2006)
Good fit is indicated by CFI greater than 0.95, RMSEA less than 0.06, nonsignificant χ2 test, and nonsignificant χ2/df values of 3 or less.
“the current rules of thumb are that good fit is indicated by CFI greater than 0.95, RMSEA less than 0.06, nonsignificant χ 2 test (i.e., p-value greater than 0.05) , and nonsignificant χ 2 /df values of 3 or less (Hoe, 2008;Hu & Bentler, 1999;Weston & Gore, 2006).”
From Evaluating Mechanisms: Mapping Weekly Dynamics between Experiential Avoidance and SAD Symptoms (2021) · cites this paper as Weston & Gore, 2006
The authors considered sample size and model complexity when judging model fit instead of solely using strict cut-offs.
“As recommended, we considered sample size and model complexity when judging model fit instead of solely using strict cut-offs, which may have led us to inappropriately reject models, given our sample was less than 500 (Weston & Gore, 2006).”
From Cognitive and behavioural factors contributing to distress in LGBTQ+students: a prospective mediation study. (2022) · cites this paper as Weston & Gore, 2006
The citation lists haven’t been fetched yet
References 70 · Cited by 2,718
- Authors
- Weston, R., & Gore, P. A.
- Affiliations
- Southern Illinois University Carbondale (US)
- ACT (US)
- Corresponding author
- Weston, Rebecca
- ORCID
- Weston, Rebecca
- Venue
- The Counseling Psychologist
- Publisher
- SAGE Publications
- Published
- 2006-09
- Volume
- 34
- Issue
- 5
- Pages
- 719-751
- ISSN
- 0011-0000 (ISSN-L)
- 1552-3861
- OpenAlex
- W2109223853
- Cited by
- 2,718
- References
- 70
- Citations by year
- 2026: 132 · 2025: 204 · 2024: 234 · 2023: 283 · 2022: 264 · 2021: 277 · 2020: 185 · 2019: 152 · 2018: 141 · 2017: 138 · 2016: 118 · 2015: 126 · 2014: 126 · 2013: 93 · 2012: 58
- FWCI
- 36.65
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- The Counseling Psychologist (publishedVersion)
- Topics
- Social Sciences › Decision Sciences › Management Science and Operations Research › Psychometric Methodologies and Testing
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
- Keywords
- structural equation modeling, model specification, system identification, data screening, model estimation, model fit assessment, model modification, sample size, model fit indices, alternative models, vocational psychology
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
