The Relative Performance of Full Information Maximum Likelihood Estimation for Missing Data in Structural Equation Models
The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.
Abstract
A Monte Carlo simulation examined the performance of 4 missing data methods in structural equation models: full information maximum likelihood (FIML), listwise deletion, pairwise deletion, and similar response pattern imputation. The effects of 3 independent variables were examined (factor loading magnitude, sample size, and missing data rate) on 4 outcome measures: convergence failures, parameter estimate bias, parameter estimate efficiency, and model goodness of fit. Results indicated that FIML estimation was superior across all conditions of the design. Under ignorable missing data conditions (missing completely at random and missing at random), FIML estimates were unbiased and more efficient than the other methods. In addition, FIML yielded the lowest proportion of convergence failures and provided near-optimal Type 1 error rates across both simulations.
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
“Models were fit using FIML estimation, which uses all available data and is the gold standard approach for handling missingness when data are missing at random or completely at random.37”
From Psychological flexibility, distress, and coping in ovarian cancer survivors. (2026) · cites this paper as [37]
In the citing paper
“The FIML algorithm generates unbiased estimation of missing data [13].”
In the citing paper
“In that case, retention of dropout cases provides more reliable model estimates than listwise deletion of dropout cases, especially when dropout rates are high (Enders & Bandalos, 2001).”
From Social media use intensity, social media use problems, and mental health among adolescents: Investigating directionality and mediating processes (2021) · cites this paper as Enders & Bandalos, 2001
In the citing paper
“Under this assumption, full information maximum likelihood (FIML) estimation yields unbiased parameter estimates by incorporating all available data (Enders & Bandalos, 2001).”
From Participation in extracurricular activities predicts academic growth and mediates socioeconomic inequalities in primary school (2026) · cites this paper as Enders & Bandalos, 2001
In the citing paper
“FIML has been shown to provide minimally biased results that are comparable to other popular missing data techniques, such as estimation maximization and multiple imputation strategies (Enders & Bandalos, 2001; Olinksi, Chen, & Harlow, 2003).”
From Perceived Parent–Child Relations, Conduct Problems, and Clinical Improvement Following the Treatment of Oppositional Defiant Disorder (2016) · cites this paper as Enders & Bandalos, 2001
3 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Maximum Likelihood from Incomplete Data Via the EM Algorithm
Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1977.tb01600.x · Cited by 50,444
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
Statistical Analysis With Missing Data
Lahiff, M., Little, R. J. A., & Rubin, D. B. (1989) · Journal of the American Statistical Association · doi:10.2307/2289883 · Cited by 17,265
RUBIN, D. B. (1976) · Biometrika · doi:10.1093/biomet/63.3.581 · Cited by 9,925
Journal of The American Statistical Association
Fienberg, S. E. (2005) · Encyclopedia of Statistical Sciences · doi:10.1002/0471667196.ess1317 · Cited by 6,125
Analysis of Incomplete Multivariate Data
Schafer, J. L. (1997) · doi:10.1201/9781439821862 · Cited by 6,012
Analysis of Incomplete Multivariate Data
Booth, D. E. (2000) · Technometrics · doi:10.1080/00401706.2000.10486013 · Cited by 5,627
Anderson, J. C., & Gerbing, D. W. (1984) · Psychometrika · doi:10.1007/bf02294170 · Cited by 1,833
(1996) · Dialnet (Universidad de la Rioja) · Cited by 1,733
Research works printed but not linked: 3
- Dempster A. P., 1977, Series B, 39, 1
- Haitovsky Y., 1968, Series B, 30, 67
- Kaplan D., 1995, Journal of Educational and Behavioral Statistics, 20, 69
Cited by
Items 201–231 of 231
Whiting, D. L., Simpson, G. K., McLeod, H. J., Deane, F. P., & Ciarrochi, J. (2012) · Brain Impairment · doi:10.1017/brimp.2012.28 · Cited by 19
Booker, J. A., Ollendick, T. H., Dunsmore, J. C., & Greene, R. W. (2016) · Journal of Child and Family Studies · doi:10.1007/s10826-015-0323-3 · Cited by 19
“FIML has been shown to provide minimally biased results that are comparable to other popular missing data techniques, such as estimation maximization and multiple imputation strategies (Enders & Bandalos, 2001; Olinksi, Chen, & Harlow, 2003).”
ResultKeough, M. T., & O'connor, R. M. (2015) · Substance Abuse: Research and Treatment · doi:10.4137/sart.s31434 · Cited by 19
Psychogiou, L., Ahun, M. N., Boivin, M., Tremblay, R. E., Orri, M., & Côté, S. M. (2022) · Development and Psychopathology · doi:10.1017/s0954579420001753 · Cited by 18
Fritz, J., Stretton, J., Askelund, A. D., Schweizer, S., Walsh, N. D., Elzinga, B. M., Goodyer, I. M., Wilkinson, P. O., & van Harmelen, A.-L. (2020) · Development and Psychopathology · doi:10.1017/s0954579419000178 · Cited by 18
Metacognitive knowledge in children at early elementary school
Haberkorn, K., Lockl, K., Pohl, S., Ebert, S., & Weinert, S. (2014) · Metacognition and Learning · doi:10.1007/s11409-014-9115-1 · Cited by 18
Tibbs, M., Deschênes, S., van der Velden, P., & Fitzgerald, A. (2025) · Journal of Youth and Adolescence · doi:10.1007/s10964-024-02093-5 · Cited by 16
“FIML is known to provide less biased estimates than pairwise or listwise deletion in structural equation models (Enders & Bandalos, 2001; Lee et al., 2019).”
Trajectories in quantitative and humanities self-efficacy during the first year of college
Han, J., Usher, E. L., & Brown, C. S. (2021) · Learning and Individual Differences · doi:10.1016/j.lindif.2021.102054 · Cited by 15
Neumann, K., Euler, H. A., Zens, R., Piskernik, B., Packman, A., St. Louis, K. O., Kell, C. A., Amir, O., Blomgren, M., Boucand, V. A., et al. (2019) · Journal of Communication Disorders · doi:10.1016/j.jcomdis.2019.105915 · Cited by 14
The role of trait mindfulness in quality of life and asthma control among adolescents with asthma
Cillessen, L., van de Ven, M. O., & Karremans, J. C. (2017) · Journal of Psychosomatic Research · doi:10.1016/j.jpsychores.2017.06.014 · Cited by 13
Konradt, U., Ellwart, T., & Gevers, J. (2021) · Learning and Individual Differences · doi:10.1016/j.lindif.2021.102003 · Cited by 11
Chan, E. S. M., & Langberg, J. M. (2026) · Journal of Occupational Rehabilitation · doi:10.1007/s10926-024-10259-y · Cited by 11
“Full information maximum likelihood (FIML) was used to estimate missing data, which is appropriate for this low level of missing data [41].”
Zhang, L., Oshri, A., Carvalho, C., Uddin, L. Q., Geier, C., Nagata, J. M., Cummins, K., Hoffman, E. A., Tomko, R. L., Chaarani, B., et al. (2025) · SLEEP · doi:10.1093/sleep/zsae234 · Cited by 10
Smith, Z. R., Del Castillo, A., Clark, O. E., & Holmbeck, G. N. (2022) · Child Neuropsychology · doi:10.1080/09297049.2022.2042503 · Cited by 10
Longitudinal and Bidirectional Relations Between Problematic Social Media Use and Cognitive Failures
Kasturiratna, K. T. A. S., & Hartanto, A. (2025) · Psychiatric Quarterly · doi:10.1007/s11126-025-10203-0 · Cited by 10
Sadikova, E., Dovgan, K., & Mazurek, M. O. (2023) · Journal of Autism and Developmental Disorders · doi:10.1007/s10803-021-05401-1 · Cited by 10
Lee, G., Hicks, D. L., Kedia, S., Regmi, S., & Mou, X. (2024) · Children and Youth Services Review · doi:10.1016/j.childyouth.2024.107456 · Cited by 9
Walderhaug, E. P., Gjestad, R., Egeland, J., Havik, O. E., & Nordgreen, T. (2019) · Nordic Journal of Psychiatry · doi:10.1080/08039488.2019.1646803 · Cited by 8
“Handling of missing data was done by Full Information Maximum Likelihood Estimation (FIML), a method which has been shown to perform better than alternative methods of handling missing data [54].”
MethodModeling trait depression amplifies the effect of childbearing on postpartum depression
Merkitch, K. G., Jonas, K. G., & O’Hara, M. W. (2017) · Journal of Affective Disorders · doi:10.1016/j.jad.2017.07.017 · Cited by 6
Acland, E. L., Pocuca, N., Paquin, S., Boivin, M., Ouellet-Morin, I., Andlauer, T. F. M., Gouin, J. P., Côté, S. M., Tremblay, R. E., Geoffroy, M., et al. (2025) · Development and Psychopathology · doi:10.1017/s095457942400004x · Cited by 6
Waters, T. E. A., Yang, R., Gu, Y., Zhu, V., Cui, L., Li, X., Way, N., Yoshikawa, H., Chen, X., Okazaki, S., et al. (2025) · Child Development · doi:10.1111/cdev.14256 · Cited by 4
Mikhail, M. E., Burt, S. A., Neale, M. C., Keel, P. K., Katzman, D. K., & Klump, K. L. (2024) · Clinical Psychological Science · doi:10.1177/21677026241230335 · Cited by 3
“…be included in the trivariate twin model even if they were missing data on the DERS because full information maximum likelihood estimation (FIML) makes use of all available data to determine model parameters, including data from cases with missing data for some predictors (Enders & Bandalos, 2001).”
Executive functions as mediators of early educational disparities by SES, gender and birth month
Rémeau, M., Schmitt, P., & Borst, G. (2026) · npj Science of Learning · doi:10.1038/s41539-026-00422-w · Cited by 3
“As the data were missing at random (MAR), we used full information maximum likelihood (FIML) estimation to handle missing data in the SEM 80 .”
InfluentialRémeau, M., Raffy, G., & Borst, G. (2026) · Journal of Youth and Adolescence · doi:10.1007/s10964-025-02281-x · Cited by 3
Goldsmith, K., Hudson, J. L., Chalder, T., Dennison, L., & Moss-Morris, R. (2020) · Behaviour Research and Therapy · doi:10.1016/j.brat.2020.103594 · Cited by 2
The epigenetic impacts of pubertal acceleration following early caregiver disruptions
Jopling, E., Drury, S. S., Fox, N. A., Zeanah, C. H., & Nelson, C. A. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2504216122 · Cited by 2
(2016) · Research Online (University of Wollongong) · Cited by 2
Zhang, W., Duan, R., Li, H., Fang, Y., Zhao, W., Wan, F., & Fan, J. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1805798
Gazelle, H., & Wu, Q. (2026) · Child Psychiatry & Human Development · doi:10.1007/s10578-026-01991-y
Psychological flexibility, distress, and coping in ovarian cancer survivors.
Telles, R., Hosch, A., Pennington, K. P., Schlumbrecht, M., McGregor, B. A., Heron, L., Zia, S., Noble, A., Goodheart, M. J., Noonan, M., et al. (2026) · Journal of Psychosocial Oncology · doi:10.1080/07347332.2025.2588632
“37 Models were estimated with a robust maximum likelihood estimator (MLR) that provides robust standard errors to account for non-normally distributed data.”
Brunet, L., Rémeau, M., Houdé, O., & Borst, G. (2026) · British Journal of Educational Psychology · doi:10.1111/bjep.70127
Metadata
- Authors
- Enders, C., & Bandalos, D.
- Affiliations
- University of Miami (US)
- University of Nebraska–Lincoln (US)
- Venue
- Structural Equation Modeling: A Multidisciplinary Journal
- Publisher
- Informa UK Limited
- Published
- 2001-07-01
- Volume
- 8
- Issue
- 3
- Pages
- 430-457
- ISSN
- 1070-5511 (ISSN-L)
- OpenAlex
- W2168646330
- Cited by
- 4,978
- References
- 27
- Citations by year
- 2026: 408 · 2025: 422 · 2024: 397 · 2023: 345 · 2022: 401 · 2021: 407 · 2020: 307 · 2019: 339 · 2018: 285 · 2017: 244 · 2016: 246 · 2015: 209 · 2014: 208 · 2013: 192 · 2012: 121
- FWCI
- 21.61
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Structural Equation Modeling A Multidisciplinary Journal (publishedVersion)
- Lincoln (University of Nebraska) (submittedVersion)
- Funders
- Mashhad University of Medical Sciences
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Decision Sciences › Management Science and Operations Research › Psychometric Methodologies and Testing
- Physical Sciences › Mathematics › Applied Mathematics › Statistical and numerical algorithms
- Keywords
- missing data, structural equation modeling, full information maximum likelihood, listwise deletion, pairwise deletion, convergence failure, model goodness of fit, missing completely at random, missing at random, sample size
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.