On Structural Equation Modeling with Data that are not Missing Completely at Random
The DOI matches its Crossref record; 44 of the 45 references it cites were found in a registry or an index.
Abstract
A general latent variable model is given which includes the specification of a missing data mechanism. This framework allows for an elucidating discussion of existing general multivariate theory bearing on maximum likelihood estimation with missing data. Here, missing completely at random is not a prerequisite for unbiased estimation in large samples, as when using the traditional listwise or pairwise present data approaches. The theory is connected with old and new results in the area of selection and factorial invariance. It is pointed out that in many applications, maximum likelihood estimation with missing data may be carried out by existing structural equation modeling software, such as LISREL and LISCOMP. Several sets of artifical data are generated within the general model framework. The proposed estimator is compared to the two traditional ones and found superior.
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 mean values of the skewness and kurtosis were smaller than the prescribed levels (skewness, 2.0 and kurtosis, 7.0), indicating no significant problems regarding multivariate normality of the data [35].”
From The Mediating Role of Generative AI Self-Regulation on Students’ Critical Thinking and Problem-Solving (2024) · cites this paper as [35]
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
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
BMDP Statistical Software 1981.
Bubolz, T. A., & Dixon, W. J. (1984) · Journal of the American Statistical Association · doi:10.2307/2288308 · Cited by 5,221
(1976) · Annals of Economic and Social Measurement · Cited by 5,035
Assessing Reliability and Stability in Panel Models
Wheaton, B., Muthen, B., Alwin, D. F., & Summers, G. F. (1977) · Sociological Methodology · doi:10.2307/270754 · Cited by 3,116
Self-citationA General Approach to Confirmatory Maximum Likelihood Factor Analysis
Jöreskog, K. G. (1969) · Psychometrika · doi:10.1007/bf02289343 · Cited by 2,489
Marginal Maximum Likelihood Estimation of Item Parameters: Application of an EM Algorithm
Bock, R. D., & Aitkin, M. (1981) · Psychometrika · doi:10.1007/bf02293801 · Cited by 2,339
Quasi-likelihood functions, generalized linear models, and the Gauss—Newton method
WEDDERBURN, R. W. M. (1974) · Biometrika · doi:10.1093/biomet/61.3.439 · Cited by 2,122 · cited 2×
“In terms of the Wheaton et al. (1977) example, an increased value of any of the socioeconomic background indicators is taken to yield an increased propensity not to attrit (a high s* value), while an increased value of the time one alienation indicators is taken to yield a decreased propensity not to attrit (i.”
BackgroundMethod
Research works printed but not linked: 5
- Little, 1983, Incomplete data in sample surveys, Vol. 2: Theory and bibliographies
- Dixon, 1983, BMDP Statistical Software
- Olsson, 1978, Selection bias in confirmatory factor analysis
- Marini, 1980, Sociological Methodology
- Lawley, 1943, A note on Karl Pearson's selection formulae, Proceedings of the Royal Society Edinburgh, Section A (Mathematics and Physics Section), 62, 28
Cited by
Items 401–500 of 881
Foster, M. E., Anthony, J. L., Zucker, T. A., & Branum-Martin, L. (2019) · Early Childhood Research Quarterly · doi:10.1016/j.ecresq.2018.02.007 · Cited by 26
Mikalsen, H. K., Bentzen, M., Säfvenbom, R., & Lagestad, P. (2020) · Frontiers in Sports and Active Living · doi:10.3389/fspor.2020.00085 · Cited by 26
“The dropout rate was handled with FIML (Muthén et al., 1987; Cham et al., 2017), and the group size is, according to Jung and Wickrama (2008) and Lubke and Neale (2006), considered sufficient.”
MethodMaltais, C., Duchesne, S., Ratelle, C. F., & Feng, B. (2015) · Learning and Individual Differences · doi:10.1016/j.lindif.2015.03.006 · Cited by 39
The development of delinquency during adolescence: a comparison of missing data techniques
Reinecke, J., & Weins, C. (2012) · Quality & Quantity · doi:10.1007/s11135-012-9721-4 · Cited by 27
Fang, H., Espy, K. A., Rizzo, M. L., Stopp, C. D., Wiebe, S. A., & Stroup, W. W. (2009) · International Journal of Information Technology & Decision Making · doi:10.1142/s0219622009003508 · Cited by 27
“42–44 These analyses can be implemented in current statistical packages such as SAS 9.1 45 and Mplus 5.0.”
InfluentialMalouf, E. T., Stuewig, J. B., & Tangney, J. P. (2012) · Addictive Behaviors · doi:10.1016/j.addbeh.2012.05.013 · Cited by 27
Factor Score Regression in Connected Measurement Models Containing Cross-Loadings
Hayes, T., & Usami, S. (2020) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2020.1729160 · Cited by 25
Liu, Y., & Sriutaisuk, S. (2021) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2021.1876520 · Cited by 25
“…the analysis and to the would-be values of this variable itself) does not hold, previous research (e.g., Asparouhov & Muthén, 2010a; Chen et al., 2020; B. Muthén et al., 1987; Shi et al., 2020) suggests that traditional missing data techniques (e.g., listwise/pairwise deletion) should be avoided.”
ResultLinking repetitive negative thinking and insomnia symptoms: A longitudinal trait-state model
Olatunji, B. O., Knowles, K. A., Cox, R. C., & Cole, D. A. (2023) · Journal of Anxiety Disorders · doi:10.1016/j.janxdis.2023.102732 · Cited by 28
When and how to use confirmatory composite analysis (CCA) in second language research
Alamer, A., Schuberth, F., & Henseler, J. (2024) · Studies in Second Language Acquisition · doi:10.1017/s0272263124000020 · Cited by 29
“Consequently, researchers can gain all the benefits they are accustomed to in SEM with latent variables, e.g., dealing with missing values (e.g., Allison, 2003; Muthén et al., 1987), gaining access to well-established model fit measures (Schermelleh-Engel et al., 2003), and constraining parameters.”
BackgroundEstimating Statistical Power With Incomplete Data
Davey, A., & Savla, J. T. (2007) · Organizational Research Methods · doi:10.1177/1094428107300366 · Cited by 31
“In order to reflect this uncertainty about their true values in the models, the conventions developed by Allison (1987) and B. Muthén et al. (1987) were used.”
BackgroundMethodKaplan, D. (1995) · Journal of Educational and Behavioral Statistics · doi:10.2307/1165388 · Cited by 25
Self-citationMaximum likelihood estimation in meta‐analytic structural equation modeling
Oort, F. J., & Jak, S. (2016) · Research Synthesis Methods · doi:10.1002/jrsm.1203 · Cited by 24
“According to Muthén et al .”
BackgroundInfluentialZhang, C., Brook, J. S., Leukefeld, C. G., & Brook, D. W. (2016) · Addictive Behaviors · doi:10.1016/j.addbeh.2016.06.019 · Cited by 24
Factor Structure and Incremental Utility of the Distress Tolerance Scale: A Bifactor Analysis
Rogers, T. A., Bardeen, J. R., Fergus, T. A., & Benfer, N. (2018) · Assessment · doi:10.1177/1073191118789496 · Cited by 27
Løvoll, H. S., Bentzen, M., & Säfvenbom, R. (2019) · Scandinavian Journal of Educational Research · doi:10.1080/00313831.2019.1639818 · Cited by 34
Shane, J., Hamm, J. M., & Heckhausen, J. (2019) · Work Aging and Retirement · doi:10.1093/workar/waz013 · Cited by 30
Maximizing data quality and shortening survey time: Three-form planned missing data survey design
Moore, E. W. G., Lang, K. M., & Grandfield, E. M. (2020) · Psychology of sport and exercise · doi:10.1016/j.psychsport.2020.101701 · Cited by 32
Liu, Y., Schuberth, F., Liu, Y., & Henseler, J. (2022) · Journal of Business Research · doi:10.1016/j.jbusres.2022.07.040 · Cited by 27
Jia, F., & Wu, W. (2022) · Behavior Research Methods · doi:10.3758/s13428-022-01936-y · Cited by 33
“Following Muthén et al. (1987), we used ±10% as the acceptable cutoff points.”
Assessing program impact using latent growth modeling: a primer for the evaluator
Hess, B. (2000) · Evaluation and Program Planning · doi:10.1016/s0149-7189(00)00032-x · Cited by 24
A Unified Approach to Multigroup Structural Equation Modeling With Nonstandard Samples
Yuan, K., & Bentler, P. M. (2001) · Psychology Press eBooks · doi:10.4324/9781410601858-7 · Cited by 24
“quadratic form of normal variates (e.g., section 1.4 of Muirhead, 1982 ).”
BackgroundInfluentialLittle, R. J. A. (2014) · Wiley eBooks · doi:10.1002/9781118445112.stat05051 · Cited by 24
An Examination of the Factor Structure of the Multidimensional Psychological Flexibility Inventory.
Thomas, K. N., Bardeen, J. R., Witte, T. K., Rogers, T. A., Benfer, N., & Clauss, K. (2022) · Assessment · doi:10.1177/10731911211024353 · Cited by 24
Klapp, A., Cliffordson, C., & Gustafsson, J. (2014) · Educational Psychology · doi:10.1080/01443410.2014.933176 · Cited by 41
Davidov, E., Thörner, S., Schmidt, P., Gosen, S., & Wolf, C. (2011) · AStA Advances in Statistical Analysis · doi:10.1007/s10182-011-0174-1 · Cited by 32
“It may be conducted by the full information maximum likelihood (FIML) procedure (Allison 1987; Muthén et al. 1987; Bollen 1989; Arbuckle 1996; Enders 2001; Brown 2006) assuming multivariate normality by maximizing logLi = Ki − 1 2 log |Σi | − 1 2 (xi − μi)′Σ−1i (xi − μi) (6) Where Li is the log…”
MethodDoes Socioeconomic Status Matter? Materialism and Self-Esteem: Longitudinal Evidence from China
Wang, R., Liu, H., & Jiang, J. (2020) · Current Psychology · doi:10.1007/s12144-020-00695-3 · Cited by 25
Lobel, M., Preis, H., Mahaffey, B., Schaal, N. K., Yirmiya, K., Atzil, S., Reuveni, I., Balestrieri, M., Penengo, C., Colli, C., et al. (2022) · Social Science & Medicine · doi:10.1016/j.socscimed.2022.115499 · Cited by 28
“For SEM, maximum likelihood estimates were used as these are robust for data missing completely at random or not (Muthén et al., 1987).”
MethodTechnology socialness and Web site satisfaction
Barnes, S. J., & Vidgen, R. T. (2014) · Technological Forecasting and Social Change · doi:10.1016/j.techfore.2014.08.017 · Cited by 30
Richter, L., Orkin, F. M., Roman, G. D., Dahly, D. L., Horta, B. L., Bhargava, S. K., Norris, S. A., & Stein, A. D. (2018) · Journal of Nutrition · doi:10.1093/jn/nxy101 · Cited by 28
“Missing data were accommodated by full information maximum likelihood, which returns unbiased parameter estimates when data are missing at random, conditional on the variables included in the model (32, 33)”
Neuroticism and sleep-onset: What is the long-term connection?
Danielsson, N. S., Jansson‐Fröjmark, M., Linton, S. J., Jutengren, G., & Stattin, H. (2009) · Personality and Individual Differences · doi:10.1016/j.paid.2009.11.023 · Cited by 26
A Note on Normal Theory Power Calculation in SEM With Data Missing Completely at Random
Dolan, C. V., van der Sluis, S., & Grasman, R. P. P. P. (2005) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1207/s15328007sem1202_4 · Cited by 27
“We consider two ways to calculate power when data are MCAR, based on Equation 1 and Equation 5. We first discuss the method based on Equation 1. As indicated by Joreskog and Sorbom (1988, p. 259; Muthen, Kaplan, & Hollis, 1987 ), in the case of MCAR, a multigroup approach may be taken, provided that the subsample asso…”
MethodWilliams, P. G., Colder, C. R., Richards, M. H., & Scalzo, C. A. (2002) · Journal of Pediatric Psychology · doi:10.1093/jpepsy/27.6.509 · Cited by 23
“Analyzing only cases with complete data has the potential to produce biased results ( Muthen, Kaplan, & Hollis, 1987 ).”
BackgroundSwitzer, F. S., & Roth, P. L. (2004) · Wiley eBooks · doi:10.1002/9780470756669.ch15 · Cited by 25
Bifactor structure and model reliability of the Test of Gross Motor Development — 3rd edition
Garn, A. C., & Webster, E. K. (2020) · Journal of science and medicine in sport · doi:10.1016/j.jsams.2020.08.009 · Cited by 25
Nichols, E., Deal, J. A., Swenor, B. K., Abraham, A. G., Armstrong, N. M., Carlson, M. C., Griswold, M., Lin, F. R., Mosley, T. H., Ramulu, P. Y., et al. (2021) · Journal of the International Neuropsychological Society · doi:10.1017/s1355617721000400 · Cited by 28
Schnoor, B., & Usanova, I. (2022) · Reading and Writing · doi:10.1007/s11145-022-10276-4 · Cited by 29
“To address missing data, we used full information maximum likelihood estimation (FIML; Dempster et al., 1977; C. K. Enders, 2010; C. Enders & Bandalos, 2001; B. O. Muthén et al., 1987; Schafer & Graham, 2002) in Mplus 8.2 software (L. K. Muthén & Muthén, 1998–2017) for model parameter estimation.”
MethodInfluentialLee, E.-H., & Kim, M. J. (2004) · Psychological Reports · doi:10.2466/pr0.95.2.659-672 · Cited by 22
Exploring Sex Differences in Worry with A Cognitive Vulnerability Model
Zalta, A. K., & Chambless, D. L. (2008) · Psychology of Women Quarterly · doi:10.1111/j.1471-6402.2008.00459.x · Cited by 22
Meta‐analytic structural equation modeling
(2015) · Wiley eBooks · doi:10.1002/9781118957813.ch7 · Cited by 23
Kuha, J., Katsikatsou, M., & Moustaki, I. (2018) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/rssa.12350 · Cited by 25
“For multiple-item data, the idea of latent response propensities appears implicitly in Lord (1974, 1983) and explicitly in Muthén et al. (1987) who stated how such models could be defined in the context of factor analysis.”
BackgroundPyrooz, D. C., Melde, C., Coffman, D. L., & Meldrum, R. C. (2021) · Criminology · doi:10.1111/1745-9125.12268 · Cited by 28
“The average relative parameter bias between a unidimensional and bifactor general factor in the data is 6.4 percent, well below the upper limit of 10 percent to 15 percent recommended by Muthén et al. (1987), suggesting no serious concern. effects with binary outcomes in MSEM (T.”
InfluentialHammer, J. H., & Brenner, R. E. (2017) · Journal of Personality Assessment · doi:10.1080/00223891.2017.1344986 · Cited by 26
Olatunji, B. O., Cox, R. C., & Cole, D. A. (2020) · Behaviour Research and Therapy · doi:10.1016/j.brat.2020.103749 · Cited by 20
Yoo, J. E. (2009) · Educational and Psychological Measurement · doi:10.1177/0013164409332225 · Cited by 25
“Following the suggestion of Muthen et al. (1987), the absolute value of bias larger than 10 was considered to indicate the estimate was biased.”
BackgroundMethodInfluentialPathways to Children's Externalizing Behavior: A Three-Generation Study
Brook, J. S., Zhang, C., Balka, E. B., & Brook, D. W. (2012) · The Journal of Genetic Psychology · doi:10.1080/00221325.2011.594821 · Cited by 21
Pagani, L. S., & Fitzpatrick, C. (2013) · Journal of Epidemiology & Community Health · doi:10.1136/jech-2012-202191 · Cited by 24
An EM Algorithm for ML Factor Analysis with Missing Data
Jamshidian, M. (1997) · Lecture notes in statistics · doi:10.1007/978-1-4612-1842-5_13 · Cited by 20
“These algorithms treat every set of observations with the same missing data pattern as a group and then impose equality constraints on the parameters across groups (see e.g., Allison, 1987; Muthen, Kaplan, and Hollis, 1987; Arminger and Sobel, 1990).”
MethodMissing by Design: Planned Missing-Data Designs in Social Science
Pokropek, A. (2011) · The Knowledge Bank (The Ohio State University) · Cited by 20
Measurement error in epidemiology: the design of validation studies II: bivariate situation
Wong, M. Y., Day, N. E., & Wareham, N. J. (1999) · Statistics in Medicine · doi:10.1002/(sici)1097-0258(19991115)18:21<2831::aid-sim282>3.0.co;2-3 · Cited by 88
Missing Data Imputation in Quality-of-Life Assessment: Imputation for WHOQOL-BREF
Lin, T. H. (2006) · PharmacoEconomics · doi:10.2165/00019053-200624090-00008 · Cited by 25
A Didactic Example of Latent Curve Analysis Applicable to the Study of Aging
Li, F., Duncan, T. E., McAuley, E., Harmer, P., & Smolkowski, K. (2000) · Journal of Aging and Health · doi:10.1177/089826430001200306 · Cited by 20
Evaluation of Hiv Risk Reduction Intervention Programs Via Latent Growth Model
Wang, J., Siegal, H. A., Falck, R. S., Carlson, R. G., & Rahman, A. (1999) · Evaluation Review · doi:10.1177/0193841x9902300604 · Cited by 20
“Muthen, Kaplan, and Hollis (1987) and Allison (1987; Allison and Hauser 1991) have developed a method by which to test and compensate for bias due to missing data.”
MethodCarmichael, J., Spitz, G., Gould, K. R., Johnston, L., Samiotis, A., & Ponsford, J. (2023) · Scientific Reports · doi:10.1038/s41598-023-35017-7 · Cited by 20
Poh, R. Y. N., Zhuang, S., Ong, X. L., & Hong, R. Y. S. (2020) · Assessment · doi:10.1177/1073191120915287 · Cited by 23
“The relative parameter bias index is used to quantify this potential bias, and a bias of between 10% and 15% is considered acceptable (B. Muthén et al., 1987).”
MethodSchmit, K. M., Brown, R., Hayer, S. D., Checovich, M. M., Gern, J. E., Wald, E. R., & Barrett, B. P. (2021) · Pediatric Research · doi:10.1038/s41390-021-01395-9 · Cited by 22
Daugherty, A. M., Sutton, B. P., Hillman, C. H., Kramer, A. F., Cohen, N. J., & Barbey, A. K. (2019) · Trends in Neuroscience and Education · doi:10.1016/j.tine.2019.100123 · Cited by 23
“In this model construction, missing values were handled via full information maximum likelihood estimation [59,60].”
MethodInfluentialTurhan, D., Scheunemann, A., Schnettler, T., Bäulke, L., Thies, D. O., Dresel, M., Fries, S., Leutner, D., Wirth, J., & Grunschel, C. (2021) · European Journal of Health Psychology · doi:10.1027/2512-8442/a000067 · Cited by 21
The preliminary development and validation of the Doping Confrontation Efficacy Scale
Sullivan, P., Feltz, D. L., LaForge-MacKenzie, K., & Hwang, S. (2014) · Psychology of sport and exercise · doi:10.1016/j.psychsport.2014.04.011 · Cited by 27
Ryder, A. B., Wilkinson, A. V., McHugh, M. K., Saunders, K. C., Kachroo, S., D'amelio, A. M., Bondy, M. L., & Etzel, C. J. (2010) · Journal of Immigrant and Minority Health · doi:10.1007/s10903-010-9415-8 · Cited by 27
“Therefore approaches that rely on the assumption of ignorability (i.e., MI) yield better results when compared with other methods such as listwise deletion [ 31 ].”
ResultAl‐Samarraie, H., Selim, H. M., & Zaqout, F. (2015) · Interactive Learning Environments · doi:10.1080/10494820.2015.1057739 · Cited by 24
“The mean values of the skewness and kurtosis were smaller than the prescribed levels (skewness: 2.0; kurtosis: 7.0), as shown in Table 3, indicating no signi fi cant problem regarding the multivariate normality of the data (Muthén, Kaplan, & Hollis, 1987).”
BackgroundInfluentialDoidge, J. (2016) · Statistical Methods in Medical Research · doi:10.1177/0962280216628902 · Cited by 19
“Maximum likelihood-based approaches to missing data, including Heckman selection models, involve explicitly modelling the likelihood of having complete data, simultaneously with implementation of the analysis model.(8,9) While having some advantages over multiple imputation and inverse probability-weighting with respe…”
MethodGustafsson, J., & Rosén, M. (2006) · Educational Research and Evaluation · doi:10.1080/13803610600697179 · Cited by 19
“As is described by Muthén, Kaplan, and Hollis (1987) such missing-data modeling can be accomplished by computing one covariance matrix for each subset of cases with a particular combination of valid observations, and by fitting a multiple-group model to all the covariance matrices, with all…”
MethodSadri, S. K., McEvoy, P. M., Pinto, A., Anderson, R., & Egan, S. J. (2018) · Journal of Personality Assessment · doi:10.1080/00223891.2018.1428983 · Cited by 19
Kopala‐Sibley, D. C., Zuroff, D. C., Hermanto, N., & Joyal‐Desmarais, K. (2015) · International Journal of Behavioral Development · doi:10.1177/0165025415573640 · Cited by 23
Beunckens, C., Molenberghs, G., Thijs, H., & Verbeke, G. N. (2007) · Statistical Methods in Medical Research · doi:10.1177/0962280206075310 · Cited by 18
Parent Involvement: Voices of Taiwanese Fathers
Hsiu-Zu, H., Tran, C. N., Ko, C.-T., Phillips, J. M., Boutin-Martinez, A. S., Dixon, C. N., & Chen, W. (2023) · International Journal about Parents in Education · doi:10.54195/ijpe.18178 · Cited by 18
Bravo, D., Oriol, X., Gómez, M. E., Cortez, D., & Unanue, W. (2020) · Frontiers in Psychology · doi:10.3389/fpsyg.2020.591498 · Cited by 22
McClintock, E. A. (2011) · Biodemography and Social Biology · doi:10.1080/19485565.2011.615172 · Cited by 25
“FIML produces less-biased estimates than listwise deletion, pairwise deletion, mean substitution, and related methods, even if the data are not missing at random (NMAR; Enders 2001 ; Muthen, Kaplan, and Hollis 1987 ), and it performs equally well as MI and Bayesian methods ( Buhi et al.”
MethodCutts, D., Johnston, R. J., Pattie, C. J., & Fisher, J. (2012) · Journal of Elections Public Opinion and Parties · doi:10.1080/17457289.2012.694888 · Cited by 23
“To obtain reliable model fits, the FIML method requires incomplete data to be at least MAR (Little & Rubin, 1987).”
MethodClauss, K. E., & Bardeen, J. R. (2018) · Journal of Personality Assessment · doi:10.1080/00223891.2018.1521417 · Cited by 19
Widaman, K. F., Grimm, K. J., Early, D. R., Robins, R. W., & Conger, R. D. (2013) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2013.797819 · Cited by 18
“Allison (1987) and Muthén et al. (1987) devised interesting ways of calculating summary matrices for multiple groups, where each group shared the same pattern of missing data on particular variables.”
BackgroundAn Estimate of the Covariance between Variables which are not Jointly Observed
Cudeck, R. A. (2000) · Psychometrika · doi:10.1007/bf02296344 · Cited by 22
Hayes, T., Baraldi, A. N., & Coxe, S. J. (2024) · Behavior Research Methods · doi:10.3758/s13428-024-02494-1 · Cited by 17
“Because participants rarely refuse to respond to questionnaire items or return to participate in subsequent study sessions for entirely random, unsystematic reasons, an MCAR mechanism should rarely be seriously entertained as plausible (Muthén et al., 1987; Raghunathan, 2004).”
Hayes, T., & McArdle, J. J. (2017) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2017.05.006 · Cited by 16
“Following prior research, we note that values of percent bias47 greater than 15 are considered problematic (Muthén et al., 1987).48 7 Another obvious point of comparison would be mice’s default method, predictive mean matching.”
Method2 Advances in Analysis of Mean and Covariance Structure when Data are Incomplete
Jamshidian, M., & Mata, M. (2007) · Handbook of computing and statistics with applications · doi:10.1016/s1871-0301(06)01002-x · Cited by 16
Factor Structure of the Spanish Version of the Ways of Coping Questionnaire
Munet‐Vilaró, F., Gregorich, S. E., & Folkman, S. (2002) · Journal of Applied Social Psychology · doi:10.1111/j.1559-1816.2002.tb00265.x · Cited by 16
Formal and Informal Model Selection with Incomplete Data
Verbeke, G., Molenberghs, G., & Beunckens, C. (2008) · Statistical Science · doi:10.1214/07-sts253 · Cited by 16
Character Education in Taiwan: A Reflection of Historical Shifts in Sociocultural Contexts
Ho, H., Lam, Y. W., & Yeh, K. (2013) · Childhood Education · doi:10.1080/00094056.2013.851590 · Cited by 18
Rosenström, T. H., Jokela, M. M., Hintsanen, M., Pulkki-Råbäck, L., Hutri‐Kähönen, N., & Keltikangas‐Järvinen, L. (2012) · Psychological Medicine · doi:10.1017/s0033291712002000 · Cited by 19
“The model was estimated with the full information maximum likelihood (FIML) approach that has been found successful in the modeling of missing data (Muthén et al. 1987), even in the case of non-normal data (Enders, 2001).”
MethodUsing Exploratory Factor Analysis for Locating Invariant Referents in Factor Invariance Studies
Finch, W. H., & French, B. F. (2008) · Journal of Modern Applied Statistical Methods · doi:10.22237/jmasm/1209615420 · Cited by 16
MAES, H. H. M., Morley, K. I., Neale, M. C., Kendler, K. S., Heath, A. C., Eaves, L. J., & Martin, N. G. (2018) · Twin Research and Human Genetics · doi:10.1017/thg.2018.22 · Cited by 18
“The fortunate consequence of maximum likelihood estimation with single relatives jointly with complete pairs is to correct the bias in mean and variance of the former toward their true population values (Little & Rubin, 1987; Muthén et al., 1987).”
BackgroundFitton, L., Hoge, R., Petscher, Y., & Wood, C. L. (2019) · Journal of Speech Language and Hearing Research · doi:10.1044/2019_jslhr-l-18-0354 · Cited by 19
Applications of Research Methodology
Scruggs, T. E., & Mastropieri, M. A. (2006) · Advances in learning and behavioral disabilities · doi:10.1016/s0735-004x(2006)19 · Cited by 21
Markov transition models for binary repeated measures with ignorable and nonignorable missing values
Yang, X., SHOPTAW, S. J., Nie, K., Liu, J., & Belin, T. R. (2007) · Statistical Methods in Medical Research · doi:10.1177/0962280206071843 · Cited by 16
Relation between work stress and job performance in nursing: A comparison of models
Steen, N., Firth, H., & Bond, S. (1998) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705519809540096 · Cited by 17
Assessing the benefits of imputing ERP projects with missing data
Myrtveit, I., Stensrud, E., & Olsson, U. H. (2002) · IEEE International Software Metrics Symposium (METRICS) · doi:10.1109/metric.2001.915517 · Cited by 17
Formal and Informal Model Selection with Incomplete Data
Verbeke, G. N., Molenberghs, G., & Beunckens, C. (2009) · Lirias · Cited by 16
Culpepper, S. A., Aguinis, H., Kern, J. L., & Millsap, R. E. (2019) · Psychometrika · doi:10.1007/s11336-018-9649-2 · Cited by 19
A multiple imputation score test for model modification in structural equation models.
Mansolf, M. A., Jorgensen, T. D., & Enders, C. K. (2019) · Psychological Methods · doi:10.1037/met0000243 · Cited by 19
Meta-analytic structural equation modelling with missing correlations
Jak, S., Oort, F. J., Roorda, D. L., & Koomen, H. M. Y. (2013) · UvA-DARE (University of Amsterdam) · Cited by 17
Trajectories of depressive symptoms in response to prolonged stress in medical interns
Guille, C., Clark, S. L., Amstadter, A. B., & Sen, S. (2013) · Acta Psychiatrica Scandinavica · doi:10.1111/acps.12137 · Cited by 19
Proportionality Assumption in Latent Basis Curve Models: A Cautionary Note
Wu, W., & Lang, K. M. (2015) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2014.938578 · Cited by 19
“Following Muthén, Kaplan, and Hollis (1987), a % relative bias that exceeds 10% is deemed to be problematic.”
BackgroundTesting a developmental model of anxiety with the Parental Facilitation of Mastery Scale
Zalta, A. K., & Chambless, D. L. (2010) · Journal of Anxiety Disorders · doi:10.1016/j.janxdis.2010.10.009 · Cited by 16
Breastfeeding and child development outcomes: an investigation of the nurturing hypothesis
Huang, J. X., Vaughn, M. G., & Kremer, K. P. (2015) · Maternal and Child Nutrition · doi:10.1111/mcn.12200 · Cited by 17
“As discussed earlier, the missing information on outcome variables (n = 349) was estimated using the Full-Information Maximum Likelihood method, a model-based solution for missing data that handles missing values and estimates model parameters in one single step (Muthén et al. 1987).”
MethodGos, E., Sagan, A., Skarzynski, P. H., & Skarżyńśki, H. (2020) · PLoS ONE · doi:10.1371/journal.pone.0237778 · Cited by 22
Family Management Practices: Research Design and Measurement Issues
Dishion, T. J., Burraston, B. O., & Li, F. (2007) · Springer eBooks · doi:10.1007/0-387-35408-5_30 · Cited by 15
Modeling Incomplete Data in Exercise Behavior Research Using Structural Equation Methodology
Duncan, T. E., Oman, R. F., & Duncan, S. C. (1994) · Journal of Sport and Exercise Psychology · doi:10.1123/jsep.16.2.187 · Cited by 15
Clauss, K. E., Witte, T. K., & Bardeen, J. R. (2021) · Journal of Personality Assessment · doi:10.1080/00223891.2021.1887879 · Cited by 14
“ARPB less than 0.10 or 0.15 suggests that the multidimensionality within a measure is not substantial enough to preclude a unidimensional solution (Muth en et al., 1987; Rodriguez et al., 2016).”
BackgroundInfluentialRocco, I., Bonati, M. R., Corso, B., & Minicuci, N. (2021) · Child Care Health and Development · doi:10.1111/cch.12874 · Cited by 20
“which has been shown to be superior to ad hoc methods for treating incomplete observations (Muthén et al., 1987) for both normal (Enders & Bandalos, 2001) and nonnormal distributions (Enders, 2001).”
Method
Metadata
- Authors
- Muthén, B., Kaplan, D., & Hollis, M.
- Affiliations
- University of California, Los Angeles (US)
- ORCID
- Muthén, Bengt
- Venue
- Psychometrika
- Publisher
- Cambridge University Press (CUP)
- Published
- 1987-09
- Volume
- 52
- Issue
- 3
- Pages
- 431-462
- ISSN
- 0033-3123 (ISSN-L)
- 1860-0980
- OpenAlex
- W2065262588
- Cited by
- 883
- References
- 45
- Citations by year
- 2026: 22 · 2025: 30 · 2024: 27 · 2023: 37 · 2022: 53 · 2021: 37 · 2020: 35 · 2019: 37 · 2018: 28 · 2017: 31 · 2016: 29 · 2015: 36 · 2014: 39 · 2013: 39 · 2012: 33
- FWCI
- 6.04
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Psychometrika (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- 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 › Applied Mathematics › Statistical and numerical algorithms
- Keywords
- structural equation modeling, missing data, missing completely at random, latent variable models, missing data mechanisms, maximum likelihood estimation, selection bias, measurement invariance, listwise deletion, pairwise deletion, LISREL
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.