Revisiting Sample Size and Number of Parameter Estimates: Some Support for the N:q Hypothesis
The DOI matches its Crossref record; 31 of the 33 references it cites were found in a registry or an index.
Abstract
A number of authors have proposed that determining an adequate sample size in structural equation modeling can be aided by considering the number of parameters to be estimated. While this advice seems plausible, little empirical support appears to exist. A previous study by Jackson (2001), failed to find support for this hypothesis, however, there were certain limitations to the study that likely led to the lack of findings. This study revisits the issue with a design modified to be more sensitive to practically significant effects of sample size to parameter estimate ratio. Consequently, some support was found for this hypothesis, notably among overall fit measures and tests. Results indicate that higher values of the observations per parameter ratio had a positive effect for some measures of fit. However, the overall effect was small relative to absolute sample size.
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
“The number of parameters is 50 in this study, so at least (50 times 20=) 1,000 participants should have been included.”
From Mediator role of presence of meaning and self-esteem in the relationship of social support and death anxiety (2022) · cites this paper as Jackson (2003)
In the citing paper
“even for latent variable models with continuous outcomes and normal distribution using ML Jackson (2003) suggested a sample-size to parameters ratio of 20:1 or at least 10:1”
From Applied Psychometrics: Sample Size and Sample Power Considerations in Factor Analysis (EFA, CFA) and SEM in General (2018) · cites this paper as Jackson, 2003
In the citing paper
“A CFA/SEM rule of thumb is the ratio of cases to free parameters, or N:q is commonly used for minimum recommendations and 10:1 to 20:1 is a commonly suggested ratio (Schumacker & Lomax, 2015; Kline, 2016; Jackson, 2003).”
From Applied Psychometrics: Sample Size and Sample Power Considerations in Factor Analysis (EFA, CFA) and SEM in General (2018) · cites this paper as Jackson, 2003
In the citing paper
“Given that the longest scale in the present study consists of 20 items (SOCS-S/SOCS-O), a minimum sample of 200 participants is required.”
From Validation of the compassionate engagement and action scales, compassion scale, and Sussex-Oxford compassion scales in a French-Canadian sample (2024) · cites this paper as [69]
In the citing paper
“In Structural Equation Modelling analyses, sample sizes of at least 10 cases per parameter are considered to be sufficient [43]. Therefore, the final sample size was satisfactory.”
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
(1998) · Cited by 36,049
Alternative Ways of Assessing Model Fit
Browne, M. W., & Cudeck, R. (1992) · Sociological Methods & Research · doi:10.1177/0049124192021002005 · Cited by 27,200
Comparative fit indexes in structural models.
Bentler, P. M. (1990) · Psychological Bulletin · doi:10.1037/0033-2909.107.2.238 · Cited by 25,266
Significance tests and goodness of fit in the analysis of covariance structures.
Bentler, P. M., & Bonett, D. G. (1980) · Psychological Bulletin · doi:10.1037//0033-2909.88.3.588 · Cited by 19,023
Structural Equations with Latent Variables.
Clogg, C. C., & Bollen, K. A. (1991) · Contemporary Sociology · doi:10.2307/2072165 · Cited by 17,442
Structural Equations with Latent Variables
Bollen, K. A. (1989) · Wiley series in probability and statistics · doi:10.1002/9781118619179 · Cited by 17,277
Power analysis and determination of sample size for covariance structure modeling.
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996) · Psychological Methods · doi:10.1037/1082-989x.1.2.130 · Cited by 10,418
Structural model evaluation and modification: An interval estimation approach
Steiger, J. H. (1990) · Multivariate Behavioral Research · doi:10.1207/s15327906mbr2502_4 · Cited by 7,408 · cited 1×
“The four absolute fit indexes examined in this study were chi-square bias, root mean squared error of approximation (RMSEA; Steiger, 1990), the Goodness-of-Fit Index (GFI; Jöreskog & Sörbom, 1986), and the centrality index (CI; McDonald, 1989).”
MethodA Reliability Coefficient for Maximum Likelihood Factor Analysis
Tucker, L. R., & Lewis, C. (1973) · Psychometrika · doi:10.1007/bf02291170 · Cited by 7,330 · cited 1×
“The Normed fit index (NFI; Bentler & Bonett, 1980), the Nonnormed fit index (NNFI; Tucker & Lewis, 1973; Bentler & Bonett, 1980), and the comparative fit index (CFI; Bentler, 1990) were used to represent Type 1, Type 2, and Type 3 indexes, respectively.”
MethodPractical issues in structural modeling
BENTLER, P. M., & CHOU, C.-P. (1987) · Sociological Methods & Research · doi:10.1177/0049124187016001004 · Cited by 5,804
(1995) · Cited by 4,008
Goodness-of-fit indexes in confirmatory factor analysis: The effect of sample size.
Marsh, H. W., Balla, J. R., & McDonald, R. P. (1988) · Psychological Bulletin · doi:10.1037//0033-2909.103.3.391 · Cited by 3,824
Evaluation of goodness-of-fit indices for structural equation models.
Mulaik, S. A., James, L. R., Van Alstine, J., & Bennett, N. (1989) · Psychological Bulletin · doi:10.1037//0033-2909.105.3.430 · Cited by 2,923
Using LISREL for Structural Equation Modeling: A Researcher′s Guide
Kelloway, E. K. (1998) · Cited by 2,789
Anderson, J. C., & Gerbing, D. W. (1984) · Psychometrika · doi:10.1007/bf02294170 · Cited by 1,833
(1995) · Cited by 1,801
Fan, X., Thompson, B., & Wang, L. (1999) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705519909540119 · Cited by 1,663
Is More Ever Too Much? The Number of Indicators per Factor in Confirmatory Factor Analysis
Marsh, H. W., Hau, K.-T., Balla, J. R., & Grayson, D. (1998) · Multivariate Behavioral Research · doi:10.1207/s15327906mbr3302_1 · Cited by 1,307
Affects of variable and subject sampling on factor pattern recovery.
Velicer, W. F., & Fava, J. L. (1998) · Psychological Methods · doi:10.1037/1082-989x.3.2.231 · Cited by 998 · cited 3×
“As noted previously, the reliability of the variables (strength of association between the variable and factor) has been found to influence CSM solutions (e.g., Anderson & Gerbing, 1984; Bandalos, 1997; Velicer & Fava, 1998).”
BackgroundMethodResultTanaka, J. S. (1987) · Child Development · doi:10.2307/1130296 · Cited by 982
Multifaceted conceptions of fit in structural equation models
Tanaka, J. S. (1993) · Medical Entomology and Zoology · Cited by 941
An evaluation of incremental fit indices: A clarification of mathematical and empirical properties.
Marsh, H. W., Balla, J. R., & Hau, K. (1996) · Dialnet (Universidad de la Rioja) · Cited by 558
Analysis of linear structural relationships by maximum likelihood and least squares methods
Jöreskog, K. G., & Sörbom, D. (1983) · Medical Entomology and Zoology · Cited by 448
Jackson, D. L. (2001) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1207/s15328007sem0802_3 · Cited by 321
Self-citationProblems in the Measurement of Latent Variables in Structural Equations Causal Models
Cohen, P., Cohen, J., Teresi, J., Marchi, M., & Velez, C. N. (1990) · Applied Psychological Measurement · doi:10.1177/014662169001400207 · Cited by 311
An index of goodness-of-fit based on noncentrality
McDonald, R. P. (1989) · Journal of Classification · doi:10.1007/bf01908590 · Cited by 306
Gerbing, D. W., & Anderson, J. C. (1985) · Multivariate Behavioral Research · doi:10.1207/s15327906mbr2003_2 · Cited by 266
Model selection in covariance structures analysis and the "problem" of sample size: A clarification.
Cudeck, R., & Henly, S. J. (1991) · Psychological Bulletin · doi:10.1037//0033-2909.109.3.512 · Cited by 239
Structural equation modeling: Back to basics
Mueller, R. O. (1997) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705519709540081 · Cited by 191
Fan, X., & Wang, L. (1998) · Educational and Psychological Measurement · doi:10.1177/0013164498058005001 · Cited by 122
Bandalos, D. L. (1997) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705519709540070 · Cited by 34
Research works printed but not linked: 4
- Bernstein, I. H. (1995, April). Simulation in teaching multivariate statistics. Paper presented at the Workshop for the Society of Applied Multivariate Research, San Antonio, TX.
- Keppel, G. (1982). Design & analysis: A researcher's handbook (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall.
- Mueller, R. O. (1996). Basic principles of SEM: An introduction to LISREL and EQS. New York: Springer.
- Ullman, J. B. (1996). Structural equation modeling. In B. G. Tabachnick & L. S. Fidell (Eds.), Using multivariate statistics (pp. 709-811). New York: HarperCollins.
Cited by
A Beginner's Guide to Structural Equation Modeling: Fourth Edition
Schumacker, R. E., & Lomax, R. G. (2015) · doi:10.4324/9781315749105 · Cited by 6,604
A Brief Guide to Structural Equation Modeling
Weston, R., & Gore, P. A. (2006) · The Counseling Psychologist · doi:10.1177/0011000006286345 · Cited by 2,718
“For example, Jackson (2001, 2003) found only a small effect of sample size on model fit when he tested the hypothesis that an inadequate sample would result in poor-fitting models.”
BackgroundStructural equations modeling: Fit Indices, sample size, and advanced topics
Iacobucci, D. (2010) · Journal of Consumer Psychology · doi:10.1016/j.jcps.2009.09.003 · Cited by 2,263
Kyriazos, T. A. (2018) · Psychology · doi:10.4236/psych.2018.98126 · Cited by 1,882
“On the other hand, even for latent variable models with continuous outcomes and normal distribution using ML Jackson (2003) suggested a sample-size to parameters ratio of 20:1 or at least 10:1.”
MethodSample Size for Survey Research: Review and Recommendations
Memon, M. A., Ting, H., Cheah, J.-H., Thurasamy, R., Chuah, F., & Cham, T. H. (2020) · Journal of Applied Structural Equation Modeling · doi:10.47263/jasem.4(2)01 · Cited by 1,343
Metadata
- Authors
- Jackson, D. L.
- Affiliations
- Yum! Brands (United States) (US)
- Corresponding author
- Jackson, Dennis L.
- Venue
- Structural Equation Modeling: A Multidisciplinary Journal
- Publisher
- Informa UK Limited
- Published
- 2003-01
- Volume
- 10
- Issue
- 1
- Pages
- 128-141
- ISSN
- 1070-5511 (ISSN-L)
- 1532-8007
- OpenAlex
- W2012802483
- Cited by
- 1,254
- References
- 33
- Citations by year
- 2026: 107 · 2025: 145 · 2024: 127 · 2023: 110 · 2022: 108 · 2021: 88 · 2020: 83 · 2019: 54 · 2018: 54 · 2017: 60 · 2016: 39 · 2015: 55 · 2014: 54 · 2013: 38 · 2012: 23
- FWCI
- 9.02
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Structural Equation Modeling A Multidisciplinary Journal (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 › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Keywords
- sample size, number of parameters, structural equation modeling, model fit measures, power analysis, model complexity
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
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.