Maximum Likelihood (ML) and Diagonally Weighted Least Squares (DWLS) Estimation Procedures: A Comparison of Estimation Bias with Ordinal and Multivariate Non-Normal Data
Open accessAbstract
This paper aims to identify the effect of using the maximum likelihood (ML) parameter estimation method when data do not meet the assumption of multivariate normality and are not continuous.Both ML and the diagonally weighted least squares (DWLS) procedure were applied to simulated sets of data, which have different distributions and include variables that can take different numbers of possible values.Results were also compared to the ideal situation of a data set consisting of continuous, normally distributed variables.Outcomes indicate that ML provides accurate results when data are continuous and uniformly distributed, but is not as precise with ordinal data that is not treated as continuous, especially when variables have a small number of categories and data do not meet the assumption of multivariate normality.In contrast, DWLS provides more accurate parameter estimates, and a model fit that is more robust to variable type and non-normality.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Confirmatory factor analysis was performed using the diagonally weighted least squares estimator because the variables for the need scale were originally categorical.
“CFA was performed using the diagonally weighted least squares estimator (DWLS), as the variables for the need scale were originally categorical variables (Mindrila, 2010).”
From Development and Validation of a Basic Psychological Needs Scale for Technology Use (2024) · cites this paper as Mindrila, 2010
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 7
- E. Babakus, C. e. Ferguson & K. G. Jöreskog, "The Sensitivity of Confirmatory Maximum Likelihood Factor Analysis to Violations of Measurement Scale and Distributional Assumptions", Journal of Marketing Research, 1987, Vol. 24, 2228.
- K. A. Bollen, Structural Equation Modeling with Latent Variables, New York: Wiley, 1989.
- M. W. Browne, "Asymptotic Distribution Free Methods in the Analysis of Covariance Structures", British Journal of Mathematical and and Statistical Psychology, 1984, Vol. 37, 62-83.
- S. J. Finney, C. A. DiStefano, Nonnormal and Categorical Data in Structural Equation Modeling, Information Age Publishing, Greenwich, Connecticut, 2006, pp. 269-314.
- J. J. Hoogland, A. Boomsma, "Robustness Studies in in Covariance Structure Modeling: An Overview and Meta-Analysis, Sociological Metods & Research, 1998, Vol. 26, pp. 329-367.
- B. O. Muthen, D. Kaplan, "A Comparison of some Methodologies for the Factor Analysis of Non- Normal Likert Variables: A Note on the Size of the Model", British Journal of Mathematical and Statistical Psychology, 1985, Vol. 45, pp. 19-30.
- R. E. Schumacker, S. T. Beyerlein, "Teacher's Corner: Confirmatory Factor Analysis With Different Correlation Types and Estimation Methods", Structural Equation Modeling: A Multidisciplinary Journal, 2000, Vol. 7, 4 , pp. 629 - 636.
Cited by
Sorted by citations
Validation and Psychometric Evaluation of the Italian Version of the Fear of COVID-19 Scale
Soraci, P., Ferrari, A., Abbiati, F. A., Del Fante, E., De Pace, R., Urso, A., & Griffiths, M. D. (2022) · International Journal of Mental Health and Addiction · doi:10.1007/s11469-020-00277-1 · Cited by 427 · see citing works
“In this specific case, since the items (see Table 1) were distributed in a non-normal way (some items out of the range of ± 1 [seeMuthén and Kaplan 1985; Tabachnick and Fidell 2007]), parallel analysis/diagonally weighted least squares method (DWLS, polychoric correlation) was used (Mindrila 2010).”
MethodUdris, R. (2014) · Computers in Human Behavior · doi:10.1016/j.chb.2014.09.036 · Cited by 234 · see citing works
Psychometric Evaluation of the Fear of COVID-19 Scale Among Chinese Population
Chi, X., Chen, S., Chen, Y., Chen, D., Yu, Q., Guo, T., Cao, Q., Zheng, X., Huang, S., Hossain, M. M., et al. (2022) · International Journal of Mental Health and Addiction · doi:10.1007/s11469-020-00441-7 · Cited by 137 · see citing works
Weigelt, O., Gierer, P., & Syrek, C. J. (2019) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph16162987 · Cited by 92 · see citing works
Bloch-Elkouby, S., Barzilay, S., Gorman, B. S., Lawrence, O. C., Rogers, M. L., Richards, J., Cohen, L. J., Johnson, B. N., & Galynker, I. (2021) · Journal of Affective Disorders · doi:10.1016/j.jad.2021.08.048 · Cited by 72 · see citing works
Islam, T., Mahmood, K., Sadiq, M., Usman, B., & Yousaf, S. U. (2020) · Frontiers in Psychology · doi:10.3389/fpsyg.2020.572526 · Cited by 54 · see citing works
“According to Mîndrilã (2010), in case of an ordinary scale, weighted least squares (WLS) parameter is best but only when data are asymmetric or show a high level of heteroskedasticity.”
BackgroundAssessment of near-term risk for suicide attempts using the suicide crisis inventory
Barzilay, S., Assounga, K., Veras, J., Beaubian, C., Bloch-Elkouby, S., & Galynker, I. (2020) · Journal of Affective Disorders · doi:10.1016/j.jad.2020.06.053 · Cited by 53 · see citing works
Measuring psychological flexibility in medical students and residents: a psychometric analysis
Palladino, C. L., Ange, B., Richardson, D. S., Casillas, R., Decker, M., Gillies, R. A., House, A., Rollock, M., Salazar, W. H., Waller, J. L., et al. (2013) · Medical Education Online · doi:10.3402/meo.v18i0.20932 · Cited by 51 · see citing works
Development and Validation of a Basic Psychological Needs Scale for Technology Use
Moradbakhti, L., Leichtmann, B., & Mara, M. (2024) · Psychological Test Adaptation and Development · doi:10.1027/2698-1866/a000062 · Cited by 41 · see citing works
Psychometric Analysis and Validation of the Italian Bergen Facebook Addiction Scale
Soraci, P., Ferrari, A., Barberis, N., Luvarà, G., Urso, A., Del Fante, E., & Griffiths, M. D. (2023) · International Journal of Mental Health and Addiction · doi:10.1007/s11469-020-00346-5 · Cited by 32 · see citing works
Cultural adaptation of ACT to support caregivers of autistic Latino children: A pilot study
Cañón, L. F., Gould, E. R., Sandoz, E. K., Moran, O., & Grimaldi, M. A. (2023) · Journal of Contextual Behavioral Science · doi:10.1016/j.jcbs.2023.03.003 · Cited by 11 · see citing works
Moran, O., & McHugh, L. (2020) · Journal of Contextual Behavioral Science · doi:10.1016/j.jcbs.2020.01.010 · Cited by 10 · see citing works
- Authors
- Mîndrilă, D.
- Affiliations
- University of South Carolina (US)
- Corresponding author
- Mîndrilă, Diana
- ORCID
- Mîndrilă, Diana
- Venue
- International Journal for Digital Society
- Publisher
- Infonomics Society
- Published
- 2010-03-01
- Volume
- 1
- Issue
- 1
- Pages
- 60-66
- ISSN
- 2040-2570 (ISSN-L)
- OpenAlex
- W4233517750
- Cited by
- 566
- References
- 0
- Citations by year
- 2026: 49 · 2025: 71 · 2024: 112 · 2023: 89 · 2022: 53 · 2021: 60 · 2020: 29 · 2019: 26 · 2018: 17 · 2017: 14 · 2016: 12 · 2015: 12 · 2014: 11 · 2013: 7 · 2012: 3
- FWCI
- 4.52
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- International Journal for Digital Society (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
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Distribution Estimation and Applications
- Keywords
- maximum likelihood estimation, estimation bias, ordinal data, categorical data, model fit, robustness, structural equation modeling
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
