Bayesian structural equation modeling: A more flexible representation of substantive theory
The DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
This article proposes a new approach to factor analysis and structural equation modeling using Bayesian analysis. The new approach replaces parameter specifications of exact zeros with approximate zeros based on informative, small-variance priors. It is argued that this produces an analysis that better reflects substantive theories. The proposed Bayesian approach is particularly beneficial in applications where parameters are added to a conventional model such that a nonidentified model is obtained if maximum-likelihood estimation is applied. This approach is useful for measurement aspects of latent variable modeling, such as with confirmatory factor analysis, and the measurement part of structural equation modeling. Two application areas are studied, cross-loadings and residual correlations in confirmatory factor analysis. An example using a full structural equation model is also presented, showing an efficient way to find model misspecification. The approach encompasses 3 elements: model testing using posterior predictive checking, model estimation, and model modification. Monte Carlo simulations and real data are analyzed using Mplus. The real-data analyses use data from Holzinger and Swineford's (1939) classic mental abilities study, Big Five personality factor data from a British survey, and science achievement data from the National Educational Longitudinal Study of 1988.
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
“This traditional approach depends on the unrealistic assumptions of zero cross-loadings [30] and limits the researchers’ ability to investigate cross-loadings and residual covariance parameters [31].”
Citations
Cites
Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003) · doi:10.1201/9780429258480 · Cited by 27,276
Alternative Ways of Assessing Model Fit
Browne, M. W., & Cudeck, R. (1992) · Sociological Methods & Research · doi:10.1177/0049124192021002005 · Cited by 27,200 · cited 1×
“BSEM with its adjoining posterior predictive p value model test is similar in spirit to the frequentist conceptualization of close fit (M. W. Browne & Cudeck, 1993).”
ResultInference from iterative simulation using multiple sequences
Gelman, A., & Rubin, D. B. (1992) · Statistical Science · doi:10.1214/ss/1177011136 · Cited by 16,627 · cited 1×
“(A8) (Appendix continues) Convergence is determined using the Gelman–Rubin convergence diagnostic (Gelman & Rubin, 1992; Gelman et al., 2004).”
MethodTesting Structural Equation Models.
Clogg, C. C., Bollen, K. A., & Long, J. S. (1995) · Social Forces · doi:10.2307/2580595 · Cited by 8,211
Markov Chain Monte Carlo in Practice
(1995) · Chapman & Hall/CRC eBooks · doi:10.1201/b14835 · Cited by 7,757
(2010) · Cited by 7,034
Analysis of Incomplete Multivariate Data
Booth, D. E. (2000) · Technometrics · doi:10.1080/00401706.2000.10486013 · Cited by 5,627 · cited 1×
“Schafer (1997) gave a statistical discussion from a missing data and multiple imputation perspective, whereas Enders (2010) gave an applied discussion of these same topics.”
BackgroundIntroduction to Statistical Mediation Analysis
MacKinnon, D. P. (2012) · doi:10.4324/9780203809556 · Cited by 5,618 · cited 1×
“An example is an indirect effect a b in a mediation model (MacKinnon, 2008).”
BackgroundHarman, H. H. (1960) · Cited by 5,148
Exploratory Structural Equation Modeling
Asparouhov, T., & Muthén, B. (2009) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705510903008204 · Cited by 2,665
Self-citation
Cited by
Items 201–300 of 1,001
A general intelligence factor in dogs
Arden, R., & Adams, M. J. (2016) · Intelligence · doi:10.1016/j.intell.2016.01.008 · Cited by 60
“We fit the CFAmodels in Mplus (Muthén &Muthén, 2012) and used a Bayesian estimator that is more sensitive to rejecting models with poor fit when sample size is small, as ours was (B. Muthén & Asparouhov, 2012).”
MethodBondt, N. D., & Petegem, P. V. (2015) · Frontiers in Psychology · doi:10.3389/fpsyg.2015.01963 · Cited by 64
Subthreshold Psychotic Symptoms in 22q11.2 Deletion Syndrome
Tang, S. X., Yi, J. J., Moore, T. M., Calkins, M. E., Kohler, C. G., Whinna, D. A., Souders, M. C., Zackai, E. H., McDonald‐McGinn, D. M., Emanuel, B. S., et al. (2014) · Journal of the American Academy of Child & Adolescent Psychiatry · doi:10.1016/j.jaac.2014.05.009 · Cited by 62
“not converge because of a Heywood case, so Bayesian estimation was used instead.(37) Composite scores were calculated using the item-grouping indicated by each of the factors and taking the mean of component subscale scores.”
Thomas, M. L. (2012) · Psychological Assessment · doi:10.1037/a0024712 · Cited by 59
Solcà, M. D. S., Bastos, L. A., Guedes, C. E. S., Bordoni, M., Borja, L. S., Larangeira, D. F., Tuy, P. G. D. S. E., Amorim, L. D. A. F., Nascimento, E. G., Oliveira, G. G. D. S., et al. (2014) · PLoS ONE · doi:10.1371/journal.pone.0103635 · Cited by 61
“The Vuong-LoMendell-Rubin likelihood ratio test was used to choose the number of classes in LCA [48].”
MethodDepression and Pain in Asian and White Americans With Knee Osteoarthritis
Ahn, H. B., Weaver, M. T., Lyon, D., Choi, E. Y., & Fillingim, R. B. (2017) · Journal of Pain · doi:10.1016/j.jpain.2017.05.007 · Cited by 64
Charisma in everyday life: Conceptualization and validation of the General Charisma Inventory.
Tskhay, K. O., Zhu, R., Zou, C., & Rule, N. O. (2017) · Journal of Personality and Social Psychology · doi:10.1037/pspp0000159 · Cited by 69
“We estimated the model using the Bayesian estimator (Muthén & Asparouhov, 2012) in Mplus (Muthén & Muthén, 2011) because it is the only estimator, to our knowledge, that can accommodate structural equation models in the context of cross-classification and boasts several other advantages.”
BackgroundMethodFeng, E., Wang, Y., & King, R. B. (2023) · Language Teaching Research · doi:10.1177/13621688221146887 · Cited by 63
“According to B. Muthén and Asparouhov (2012), the smaller values of BIC, aBIC, and AIC indicate a better model fit, while a significant L-M-R LRT and BLRT indicate that the K model is better than the K-1 model.”
Cooley, J. L., Fite, P. J., & Pederson, C. A. (2017) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-017-0283-8 · Cited by 60
“findings and yield parameters that coincide with maximum likelihood estimates (Muthén and Asparouhov 2012; for a review of Bayesian estimation, see Zyphur and Oswald 2015).”
BackgroundMethodMiddle school engagement profiles: Implications for motivation and achievement in science
Bae, C. L., & DeBusk-Lane, M. (2019) · Learning and Individual Differences · doi:10.1016/j.lindif.2019.101753 · Cited by 59
Abramovitch (2019) · doi:10.1016/j.cpr.2018.09.001 · Cited by 59
Psychometrics in Behavioral Software Engineering: A Methodological Introduction with Guidelines
Graziotin, D., Lenberg, P., FELDT, R. H., & Wagner, S. (2021) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3469888 · Cited by 68
“[66], Muthén and Asparouhov [71].”
Cui, Z., Pines, A. R., Larsen, B., Sydnor, V. J., Li, H., Adebimpe, A., Alexander‐Bloch, A. F., Bassett, D. S., Bertolero, M. A., Calkins, M. E., et al. (2022) · Biological Psychiatry · doi:10.1016/j.biopsych.2022.05.014 · Cited by 66
“The confirmatory bifactor model was estimated using a Bayesian estimator in Mplus (19).”
MethodWang, J., Hefetz, A., & Liberman, G. (2017) · Culture and Education · doi:10.1080/11356405.2017.1367907 · Cited by 63
“Por ejemplo, para la estimación de modelos con muestras pequeñas puede usarse el enfoque bayesiano, el cual no depende de la teoría asintótica (Lee & Song, 2004; Muthén & Asparouhov, 2012).”
BackgroundMethodSanz, J. H., Wang, J., Berl, M. M., Armour, A. C., Cheng, Y. I., & Donofrio, M. T. (2018) · The Journal of Pediatrics · doi:10.1016/j.jpeds.2018.07.018 · Cited by 61
Yu, S., & Levesque-Bristol, C. (2020) · Contemporary Educational Psychology · doi:10.1016/j.cedpsych.2020.101857 · Cited by 58
“…calls for the use of cross-classified models to account for multiple nesting structures and to obtain more accurate estimations (e.g., Asparouhov & Muthén, 2012; Dunn, Richmond, Milliren, & Subramanian, 2015; Marsh, Martin, & Cheng, 2008), we have found no prior study that uses a cross-classified…”
Li, C., Ivarsson, A., Lam, L. T.-M., & Sun, J. (2019) · Frontiers in Psychology · doi:10.3389/fpsyg.2019.00665 · Cited by 67
“50 together with its 95% confidence intervals centering 0 are considered as an indication of good model fit (Muthén and Asparouhov, 2012).”
BackgroundNájera, P., Abad, F. J., & Sorrel, M. Á. (2023) · Psychological Methods · doi:10.1037/met0000579 · Cited by 67
Dietrich, J., Shulman, S., & Nurmi, J. (2013) · Journal of Research in Personality · doi:10.1016/j.jrp.2013.06.004 · Cited by 60
Lu, Y., Zhang, M. M., Yang, M. M.-H., & Li, T. (2024) · Human Resource Management · doi:10.1002/hrm.22270 · Cited by 59
Chan, D. K. C., Ivarsson, A., Stenling, A., Yang, S. X., Chatzisarantis, N. L. D., & Hagger, M. S. (2015) · Journal of Sport and Exercise Psychology · doi:10.1123/jsep.2015-0045 · Cited by 55
Birgisdóttir, F., Gestsdóttir, S., & Þórsdóttir, F. (2015) · Early Education and Development · doi:10.1080/10409289.2015.1003505 · Cited by 64
“The Bayesian estimator was used because of its ability to handle small sample sizes, missing data, strongly skewed distributions, and few clusters (Muthén & Asparouhov, 2010, 2012).”
MethodDvorak, R. D., Pearson, M. R., Neighbors, C., & Martens, M. P. (2015) · Journal of Consulting and Clinical Psychology · doi:10.1037/a0038902 · Cited by 57
“The analyses utilized a Bayesian approach, which is appropriate for nested/hierarchical data (Dagne, Howe, Brown, & Muthén, 2002) and analytically robust with small samples (B. O. Muthén & Asparouhov, 2012).”
Fong, T. C. T., & Ho, R. (2013) · Quality of Life Research · doi:10.1007/s11136-013-0429-2 · Cited by 55
“Based on the Bayesian structural equation modeling (BSEM) approach [24, 25], Muthén and Asparouhov [21, 26] recently pioneered a new statistical approach in CFA and SEM studies.”
BackgroundMethodInfluentialPasha, O. Q., Poister, T. H., Wright, B. E., & Thomas, J. C. (2017) · Public Performance & Management Review · doi:10.1080/15309576.2017.1335220 · Cited by 66
“PPP is obtained via an F-statistic based on the usual likelihood-ratio χ2 test, comparing the null hypothesis against an unrestricted hypothesis (Muthén & Asparouhov, 2012).”
MethodA penalized likelihood method for multi‐group structural equation modelling
Huang, P. (2018) · British Journal of Mathematical and Statistical Psychology · doi:10.1111/bmsp.12130 · Cited by 62
Imperatori, C., Fabbricatore, M., Lester, D., Manzoni, G. M., Castelnuovo, G., Raimondi, G., & Innamorati, M. (2018) · Eating and Weight Disorders - Studies on Anorexia Bulimia and Obesity · doi:10.1007/s40519-018-0607-x · Cited by 55
“A nonsignificant PPP indicated the fit of the model, while the sensitivity analysis suggested a prior data conflict, with more stringent prior variance being associated with wider Bayesian credibility intervals.”
MethodInfluentialEnvironmental exposure to manganese in air: Associations with tremor and motor function
Bowler, R. M., Beseler, C. L., Gocheva, V. V., Colledge, M., Kornblith, E. S., Julian, J. R., Kim, Y., Bollweg, G. L., & Lobdell, D. T. (2015) · The Science of The Total Environment · doi:10.1016/j.scitotenv.2015.09.084 · Cited by 54
“Ideally, the PPV should be about 0.5 and should be centered at 0 with a narrow and symmetric 95% confidence interval, indicating an excellent fitting model (Muthén and Asparouhov, 2012).”
BackgroundInfluentialCross-cultural measurement invariance of the Eysenck Personality Questionnaire across 33 countries
Bowden, S., Saklofske, D., van de Vijver, F., Sudarshan, N. J., & Eysenck, S. (2016) · Personality and Individual Differences · doi:10.1016/j.paid.2016.04.028 · Cited by 47
Challenging Conventional Wisdom for Multivariate Statistical Models With Small Samples
McNeish, D. (2017) · Review of Educational Research · doi:10.3102/0034654317727727 · Cited by 57
“…properties at smaller samples because they yield more accurate estimates of sampling variability compared to frequentist methods, which consequently leads to more appropriate statistical inferences (Depaoli & van de Schoot, 2017; Dunson, 2001; Lee & Song, 2004; Muthén & Asparouhov, 2012).”
BackgroundMethodEngagement profiles of elementary students in urban schools
Bae, C. L. L., DeBusk‐Lane, M., & Lester, A. M. (2020) · Contemporary Educational Psychology · doi:10.1016/j.cedpsych.2020.101880 · Cited by 57
Lehmann, S., Breivik, K., Heiervang, E. R., Havik, T., & Havik, O. E. (2015) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-015-0045-4 · Cited by 62
“…analyses we tested the robustness of the results of the CFA analyses for the three models, by rerunning the analyses with a Bayesian estimator (with uninformative priors) as this estimator is less sensitive for small sample sizes (B. Muthén and Asparouhov 2012; van de Schoot et al. 2014).”
BackgroundMethodInfluentialHogue, A., Bobek, M., Dauber, S., Henderson, C. E., McLeod, B. D., & Southam-Gerow, M. A. (2019) · Journal of Clinical Child & Adolescent Psychology · doi:10.1080/15374416.2018.1555762 · Cited by 53
“Specifically, Muthén and Asparouhov (2012) advocated using prior distributions with small variances for cross-loading factors and residual covariances.”
BackgroundBardach, L., Yanagida, T., Goetz, T., Jach, H. K., & Pekrun, R. H. (2023) · Developmental Psychology · doi:10.1037/dev0001537 · Cited by 61
“A PPP value close to .50 and a χ 2 difference of 0 falling close to the middle of the confidence interval imply good fit (Muthén & Asparouhov, 2012).”
BackgroundEmotional Distress, Alcohol Use, and Bidirectional Partner Violence Among Lesbian Women
Lewis, R. J., Padilla, M. A., Milletich, R. J., Kelley, M. L., Winstead, B. A., Lau‐Barraco, C., & Mason, T. B. (2015) · Violence Against Women · doi:10.1177/1077801215589375 · Cited by 55
“Model fit was assessed using posterior predictive checking with the following criteria: posterior predictive p value > .10 and a chi-square difference test between the observed and replicated data CI that captures zero (Muthén & Asparouhov, 2012).”
A comparison of Bayesian and frequentist model selection methods for factor analysis models.
Lu, Z., Chow, S., & Loken, E. (2017) · Psychological Methods · doi:10.1037/met0000145 · Cited by 57
“…the context of exploratory FA (EFA) such as rotational or identification constraints (Jennrich & Sampson, 1966), and Bayesian approaches to handling these issues (Lu et al., 2016; Mavridis & Ntzoufras, 2014; Muthén & Asparouhov, 2012) are beyond the scope of this article and are not addressed here.”
BackgroundMethodResultInfluentialHow, M.-L., & Hung, D. W. L. (2019) · Education Sciences · doi:10.3390/educsci9020110 · Cited by 54
“The Bayesian approach was utilized by researchers in education, such as Kaplan [21], Levy [22], Mathys [23], and Muthén and Asparouhov [24].”
MethodOwner of a lonely heart: The stability of loneliness across the life span.
Mund, M., Lüdtke, O., & Neyer, F. J. (2019) · Journal of Personality and Social Psychology · doi:10.1037/pspp0000262 · Cited by 59
Item selection, evaluation, and simple structure in personality data
Pettersson, E., & Turkheimer, E. N. (2010) · Journal of Research in Personality · doi:10.1016/j.jrp.2010.03.002 · Cited by 62
Fong, T. C. T., Chan, J. S. M., Chan, C., Ho, R., Ziea, E. T.-C., Wong, V. C. W., Ng, B. F. L., & Ng, S. (2015) · Quality of Life Research · doi:10.1007/s11136-015-0944-4 · Cited by 55
“Traditional CFA has been criticized for being overly restrictive in fixing all cross-loadings to zero [14].”
BackgroundPsychometric Properties of the CES-D Among Black Adolescents in Public Housing
Lu, W., Lindsey, M. A., Irsheid, S. B., & Nebbitt, V. E. (2017) · Journal of the Society for Social Work and Research · doi:10.1086/694791 · Cited by 55
“Based on theory or prior analyses, traditional CFA specifies a number of factors with cross-loadings fixed at zero to reflect a hypothesis that each indicator is influenced by a single factor (Muthén & Asparouhov, 2012).”
BackgroundMethodShould Regularization Replace Simple Structure Rotation in Exploratory Factor Analysis?
Scharf, F., & Nestler, S. (2019) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2018.1558060 · Cited by 65
Morbée, S., Vansteenkiste, M., Aelterman, N., & Haerens, L. L. (2020) · The Sport Psychologist · doi:10.1123/tsp.2018-0197 · Cited by 58
“An excellent fitting model is expected to have a PPP-value around 0.5 266 (Muthén & Asparouhov, 2012).”
BackgroundInfluentialStronge, S., Bulbulia, J. A., Davis, D. E., & Sibley, C. G. (2020) · Social Psychological and Personality Science · doi:10.1177/1948550620942381 · Cited by 58
Data fusion for estimating Macroscopic Fundamental Diagram in large-scale urban networks
Saffari, E., Yildirimoḡlu, M., & Hickman, M. D. (2022) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2022.103555 · Cited by 52
Lan, H., Ma, X., Qiao, W., & Liu, Y. (2022) · Ocean & Coastal Management · doi:10.1016/j.ocecoaman.2022.106189 · Cited by 52
Zhang, L., Wei, X., Lu, J., & Pan, J. (2020) · Advances in Psychological Science · doi:10.3724/sp.j.1042.2020.01777 · Cited by 52
Cheng, C. C., Elsworth, G. R., & Osborne, R. H. (2021) · Journal of Medical Internet Research · doi:10.2196/30243 · Cited by 55
“A prior variance of 0.01 or 0.02 means that 95% of the variation lies within the ranges of ±0.20 or ±0.28 [81,82].”
InfluentialTowards a behavioral software engineering
Lenberg, P., FELDT, R. H., & Wallgren, L. G. (2014) · International Workshop on Cooperative and Human Aspects of Software Engineering (CHASE) · doi:10.1145/2593702.2593711 · Cited by 61
“alternatives have been proposed [49, 31, 57, 13].”
Liang, X., & Jacobucci, R. (2019) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2019.1693273 · Cited by 53
Heinzel, S., Tschorn, M., Schulte-Hutner, M., Schäfer, F., Reese, G., Pohle, C., Peter, F., Neuber, M., Liu, S., Keller, J., et al. (2023) · Frontiers in Psychology · doi:10.3389/fpsyg.2023.1239425 · Cited by 50
“We compared a four-factorial model in which the main loadings of the items on their hypothesized factors were specified to a fourfactorial model in which near-zero cross-loadings of items from the three factors rumination, behavioral symptoms and anxiety about personal impact were allowed on the affective symptoms fac…”
MethodA diary study on shared leadership, team work engagement, and goal attainment
Klasmeier, K. N., & Rowold, J. (2021) · Journal of Occupational and Organizational Psychology · doi:10.1111/joop.12371 · Cited by 58
Lopes, V. P., Barnett, L. M., Saraiva, L., Gonçalves, C., Bowe, S. J., Abbott, G., & Rodrigues, L. P. (2016) · Child Care Health and Development · doi:10.1111/cch.12359 · Cited by 50
Effects of a rubric for mathematical reasoning on teaching and learning in primary school
Smit, R., Bachmann, P., Blum, V., Birri, T., & Hess, K. (2017) · Instructional Science · doi:10.1007/s11251-017-9416-2 · Cited by 32
“switched to Bayesian estimation, which allows for better model estimation because largesample theory is not needed (Muthén and Asparouhov 2012).”
BackgroundMethodKlasmeier, K. N., Schleu, J. E., Millhoff, C., Poethke, U., & Bormann, K. C. (2021) · European Journal of Work and Organizational Psychology · doi:10.1080/1359432x.2021.1968375 · Cited by 56
“PPP was close to the ideal value of .5, (PPP = .439) and the PPC confidence interval included zero (PPC 95% CI using chi-square = [−20.544; 23.591]; B. O. Muthén & Asparouhov, 2012).”
BackgroundMethodInfluentialHopes and cautions in implementing Bayesian structural equation modeling.
MacCallum, R. C., Edwards, M. C., & Cai, L. (2012) · Psychological Methods · doi:10.1037/a0027131 · Cited by 50
“We agree with Muthén and Asparouhov (2012) about a number of potential benefits of BSEM compared with conventional likelihood-based SEM.”
BackgroundMethodInfluentialThrash, T. M., Maruskin, L. A., Moldovan, E. G., Oleynick, V. C., & Belzak, W. C. M. (2016) · Journal of Personality and Social Psychology · doi:10.1037/pspp0000094 · Cited by 55
Lecuona, O., García-Garzón, E., García-Rubio, C., & Rodríguez-Carvajal, R. (2020) · Assessment · doi:10.1177/1073191119873718 · Cited by 45
“To do so, cross-loadings are given small variance prior distributions, which reflect the researcher’s expectations for these parameters to be close to Lecuona et al. 863 zero in the final solution (Asparouhov, Muthén, & Morin, 2015; Muthén & Asparouhov, 2012).”
BackgroundMethodCultural threat perceptions predict violent extremism via need for cognitive closure
Obaidi, M., Anjum, G., Bierwiaczonek, K., Dovidio, J. F., Ozer, S., & Kunst, J. R. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2213874120 · Cited by 51
A Bifactor Approach to Model Multifaceted Constructs in Statistical Mediation Analysis
González, O., & MacKinnon, D. P. (2016) · Educational and Psychological Measurement · doi:10.1177/0013164416673689 · Cited by 51
“Exploratory (Asparouhov & Muthén, 2009; Morin et al., 2016) and Bayesian (Muthén & Asparouhov, 2012) approaches to bifactor modeling could help reduce convergence problems due to low factor loadings, small sample sizes, or the orthogonality of specific factors.”
BackgroundMatsuzaka, C. T., Wainberg, M. L., Pala, A. N., Hoffmann, E. V., Coimbra, B. M., Braga, R. F., Sweetland, A. C., & Mello, M. F. (2017) · BMC Psychiatry · doi:10.1186/s12888-017-1379-y · Cited by 55
“Given the relatively small sample size, we tested group differences on primary and secondary outcomes using Latent Growth Model (LGM) and Bayesian estimator [59].”
MethodMeasurement invariance of the Childhood Autism Rating Scale ( CARS ) across six countries
Stevanović, D., Costanzo, F., Fucà, E., Valeri, G., Vicari, S., Robins, D. L., Samms‐Vaughan, M. E., Erkuran, H. Ö., Yaylacı, F., Deshpande, S. N., et al. (2021) · Autism Research · doi:10.1002/aur.2586 · Cited by 52
Are conservative approaches to new product selling a blessing in disguise?
van der Borgh, M., & Schepers, J. (2017) · Journal of the Academy of Marketing Science · doi:10.1007/s11747-017-0521-1 · Cited by 50
“Given these conditions we employed Bayesian methods because these provide more reliable estimations on small samples (Muthén and Asparouhov 2012) and do not assume or require normal distributions for the model parameters (Zhang et al. 2009).”
MethodInfluentialMental toughness and behavioural perseverance: A conceptual replication and extension
Giles, B., Goods, P. S. R., Warner, D. R., Quain, D., Peeling, P., Ducker, K. J., DAWSON, B. T., & Gucciardi, D. F. (2017) · Journal of science and medicine in sport · doi:10.1016/j.jsams.2017.10.036 · Cited by 47
A Guide to Detecting and Modeling Local Dependence in Latent Class Analysis Models
Visser, M., & Depaoli, S. A. (2022) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2022.2033622 · Cited by 46
Lüdtke, O., Robitzsch, A., & Wagner, J. (2017) · Psychological Methods · doi:10.1037/met0000155 · Cited by 51
Questioning Psychological Constructs: Current Issues and Proposed Changes
de Boeck, P. A. L., Pek, J. C. X., Walton, K. M., Wegener, D. T., Turner, B. M., Andersen, B. L., Beauchaine, T. P., Lecavalier, L., Myung, J. I., & Petty, R. E. (2023) · Psychological Inquiry · doi:10.1080/1047840x.2023.2274429 · Cited by 56
“…solution, which are less strictly confirmatory (e.g., Bayesian approaches with small cross-loadings and small violations of measurement invariance, as presented by Muth � en and Asparouhov (2012) and Asparouhov and Muth � en (2014), are counter-indications of treating constructs as points in space.”
Body image disturbance and health behaviors among sexual minority men living with HIV.
Blashill, A. J., Goshe, B. M., Robbins, G. K., Mayer, K. H., & Safren, S. A. (2014) · Health Psychology · doi:10.1037/hea0000081 · Cited by 45
“See Muthén and Asparouhov (2012) for a thorough review of Bayesian approaches of testing structural models.”
BackgroundStenling, A., Ivarsson, A., Hassmén, P., & Lindwall, M. (2017) · Psychology of sport and exercise · doi:10.1016/j.psychsport.2017.03.002 · Cited by 51
“A well-fitting model should have a PPp value of around 0.50 and a symmetric 95% CI centering on zero (Levy, 2011; Muth en & Asparouhov, 2012).”
BackgroundInfluentialBayesian SEM for Specification Search Problems in Testing Factorial Invariance
Shi, D., Song, H., Liao, X., Terry, R., & Snyder, L. A. (2017) · Multivariate Behavioral Research · doi:10.1080/00273171.2017.1306432 · Cited by 45
“They often reflect strong substantive beliefs about the parameters, and can be decided based on theory, knowledge of experts, or results fromprevious studies (Muthén & Asparouhov, 2012a; Kaplan & Depaoli, 2012).”
BackgroundMethodInfluentialLomazzi, V. (2018) · Methoden, Daten, Analysen (mda) · doi:10.12758/mda.2017.09 · Cited by 58
“Nowadays, the concept of “approximate equivalence” introduced by Muthén and Asparouhov (2012, 2013), appears to be the most feasible way of navigating between the two mythological monsters.”
BackgroundMomentary precipitants connecting stress and smoking lapse during a quit attempt.
Cambron, C., Haslam, A. K., Baucom, B., Lam, C. Y., Vinci, C., Cinciripini, P. M., Li, L., & Wetter, D. W. (2019) · Health Psychology · doi:10.1037/hea0000797 · Cited by 54
“To determine significance, the 95% highest probability density interval (HPD) was calculated from the posterior distribution of the model and CIs based on the HPD that did not include zero were considered statistically different from zero (Muthén & Asparouhov, 2012).”
MethodInfluentialSelf-set goals improve academic performance through nonlinear effects on daily study performance
Alessandri, G., Боргогни, Л., Latham, G. P., Cepale, G., Theodorou, A., & Longis, E. D. (2019) · Learning and Individual Differences · doi:10.1016/j.lindif.2019.101784 · Cited by 52
Janssen, E. P. C. J., de Vugt, M. E., Köhler, S., Wolfs, C. A. G., Kerpershoek, L., Handels, R. L. H., Orrell, M. W., Woods, B. T., Jelley, H., Stephan, A., et al. (2016) · Aging & Mental Health · doi:10.1080/13607863.2016.1255716 · Cited by 49
“…model was identified, based on comparison of Bayesian Information Criterion (lower is better), classification entropy (preferably .80) and Lo–Mendel–Rubin adjusted likelihood ratio testing comparing models with k classes to the model with k ¡ 1 classes (Muthen & Asparouhov, 2012; Schwarz, 1978).”
MethodAn initial analysis of software engineers’ attitudes towards organizational change
Lenberg, P., Tengberg, L. G. W., & FELDT, R. H. (2016) · Empirical Software Engineering · doi:10.1007/s10664-016-9482-0 · Cited by 57
“In particular, Muthén and Asparouhov’s (Muthén and Asparouhov 2012) Bayesian SEM approach replaces the parameter specification of exact zeros and exact equalities with approximate zeros and equalities‘ and can thus better represent substantive theories without multiple, often ad hoc, model rejection and retry steps as…”
MethodBayesian Estimation of Single-Test Reliability Coefficients
Pfadt, J. M., van den Bergh, D., Sijtsma, K., Moshagen, M., & Wagenmakers, E. (2021) · Multivariate Behavioral Research · doi:10.1080/00273171.2021.1891855 · Cited by 48
“For this, one needs to apply a more complex procedure of posterior sampling than the single-factor model (see e.g., Lee, 2007; Muth (cid:4) en & Asparouhov, 2012).”
The subjective career success of women: The role of personal resources
Koekemoer, E., Olckers, C., & Schaap, P. (2023) · Frontiers in Psychology · doi:10.3389/fpsyg.2023.1121989 · Cited by 49
“10) could be identified, which suggests the lowerthan-ideal CFI and TLI values can be ascribed to multiple non-substantive misspecifications or white noise (Muthén and Asparouhov, 2012).”
BackgroundHagger, M. S., & Hamilton, K. (2018) · Learning and Individual Differences · doi:10.1016/j.lindif.2018.09.002 · Cited by 50
Roover, K. D. (2021) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2020.1866577 · Cited by 46
“Moreover, fixed zero loadings are often too restrictive (Asparouhov & Muthén, 2009; McCrae et al., 1996; Muthén & Asparouhov, 2012) and model misspecifications can severely bias other parameter estimates (Anderson & Gerbing, 1982; Bollen et al., 2007).”
BackgroundSport-to-School Spillover Effects of Passion for Sport: The Role of Identity in Academic Performance
Yukhymenko–Lescroart, M. A. (2021) · Psychological Reports · doi:10.1177/00332941211006925 · Cited by 44
Fischer, J., Praetorius, A., & Klieme, E. (2019) · Educational Assessment Evaluation and Accountability · doi:10.1007/s11092-019-09295-7 · Cited by 47
“…invariance, see Byrne et al. 1989), allowing small crosscountry parameter differences (e.g., Bayesian approximate invariance testing, see B.O. Muthén and Asparouhov 2012), or favoring a model with most invariant and a minimum of non-invariant parameters (e.g., Alignment, see Asparouhov and…”
BackgroundBifactor exploratory structural equation modeling: A meta-analytic review of model fit
Gegenfurtner, A. (2022) · Frontiers in Psychology · doi:10.3389/fpsyg.2022.1037111 · Cited by 48
Using Hamiltonian Monte Carlo to estimate the log-linear cognitive diagnosis model via Stan
Jiang, Z., & Carter, R. A. (2018) · Behavior Research Methods · doi:10.3758/s13428-018-1069-9 · Cited by 49
“The core of Bayesian inferences, Markov chain Monte Carlo (MCMC) techniques can provide accurate estimation for models with high complexity, where many traditional methods cannot (see, e.g., Muthén & Asparouhov, 2012).”
MethodQuantitative research in tourism and hospitality: an agenda for best-practice recommendations
Assaf, A. G., & Tsionas, E. G. (2019) · International Journal of Contemporary Hospitality Management · doi:10.1108/ijchm-02-2019-0148 · Cited by 49
Feng, Z., Wang, L., Peng, Q., Li, J., & Liang, T. (2022) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2022.131084 · Cited by 43
Ozechowski, T. J. (2014) · Journal of Consulting and Clinical Psychology · doi:10.1037/a0035889 · Cited by 42
“Draws from the posterior distribution, simulated using MCMC sampling methods, are thereby restricted to the same space (Chung et al., 2011; Gelman, 2006; Muthén & Asparouhov, 2012; Skrondal & RabeHesketh, 2004, p. 206).”
InfluentialJason, L. A., Guerrero, M., Salomon‐Amend, M., Stevens, E., Light, J. M., & Stoolmiller, M. (2021) · American Journal of Community Psychology · doi:10.1002/ajcp.12481 · Cited by 42
Stromeyer, W. R., Miller, J. W., Sriramachandramurthy, R., & DeMartino, R. (2014) · Journal of Management · doi:10.1177/0149206314551962 · Cited by 51
“In contrast, with the BSEM approach, the potential model changes that we observe are estimated jointly in a single analysis (Muthén & Asparouhov, 2012a).”
BackgroundMethodInfluentialValidity, Reliability, and Factor Structure of the Secondary Traumatic Stress Scale-French Version
Jacobs, I., Charmillot, M., Martin-Söelch, C., & Horsch, A. (2019) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2019.00191 · Cited by 46
“As a consequence, BSEM with informative small-variance priors shrinks the crossloadings toward their zero prior mean likely yielding smaller cross-loadings than ESEM with target rotation (57).”
MethodActual and perceived motor competence: Are children accurate in their perceptions?
Morano, M., Bortoli, L., Ruiz, M. C., Campanozzi, A., & Robazza, C. (2020) · PLoS ONE · doi:10.1371/journal.pone.0233190 · Cited by 42
“Muthén and Asparouhov [31] advocated Bayesian analysis as an alternative approach to factor analysis and structural equation modeling (SEM) frequentist methods.”
BackgroundMethodInfluentialGarcés, C. R., Fuentes, G. P., & Fuentes, M. A. G. (2020) · Cuadernos de Investigación Educativa · doi:10.18861/cied.2020.11.2.2995 · Cited by 42
Moen, F., Vatn, M., Olsen, M., Haugan, J. A., & Skalická, V. (2022) · Frontiers in Sports and Active Living · doi:10.3389/fspor.2021.697535 · Cited by 46
“Bayesian estimator uses all information available and provides estimates unbiased by missingness (Muthén and Asparouhov, 2012; Schuurman et al., 2016; Asparouhov and Muthén, 2020).”
MethodComparing sport motivation scales: A response to Pelletier et al.
Lonsdale, C., Hodge, K. P., Hargreaves, E. A., & Ng, J. Y. Y. (2014) · Psychology of sport and exercise · doi:10.1016/j.psychsport.2014.03.006 · Cited by 49
“…CFA, models that allow for small cross-loadings, such as ones adopting exploratory structural equation modeling (Asparouhov & Muthén, 2009) or Bayesian structural equation modeling (Muthén & Asparouhov, 2012) frameworks, may be used to address the strong factor correlations found in some studies.”
Slesnick, N., Zhang, J., & Brakenhoff, B. R. (2016) · Children and Youth Services Review · doi:10.1016/j.childyouth.2016.11.033 · Cited by 43
Bei, B., Wiley, J. F., Allen, N. B., & Trinder, J. (2015) · Sleep · doi:10.5665/sleep.4508 · Cited by 37
Johnson, A. R., van de Schoot, R., Delmar, F., & Crano, W. D. (2014) · Journal of Management · doi:10.1177/0149206314539351 · Cited by 41
“With uninformative priors, Bayesian estimation is more likely to converge and can provide results that are comparable with ML, had it converged (B. O. Muthén & Asparouhov, 2012).”
ResultLee, R. S. C., Hermens, D. F., Scott, J. L., O’Dea, B., Glozier, N., Scott, E. M., & Hickie, I. B. (2017) · Psychological Medicine · doi:10.1017/s0033291717000484 · Cited by 41
“The posterior distribution of Bayesian estimation was achieved through an iterative Markov chain Monte Carlo (MCMC) algorithm using the Gibbs sampler method (Muthen & Asparouhov, 2012).”
MethodSantirso, F. A., Martín-Fernández, M., Lila, M., Gracia, E., & Terreros, E. (2018) · International Journal of Clinical and Health Psychology · doi:10.1016/j.ijchp.2018.02.003 · Cited by 41
Griffith, G. J., & Jones, K. (2019) · Social Science & Medicine · doi:10.1016/j.socscimed.2019.112638 · Cited by 45
Testing the Mindfulness-to-Meaning Theory in Daily Life
Sgherza, T. R., DeMarree, K. G., & Naragon‐Gainey, K. (2022) · Mindfulness · doi:10.1007/s12671-022-01961-0 · Cited by 45
“Non-informative priors were used to freely estimate parameters, given that informative priors need to be generated from results of previous studies and there are no within-person studies of this model to draw from (see Muthen & Asparouhov, 2012).”
BackgroundMethod
Metadata
- Authors
- Muthén, B., & Asparouhov, T.
- Affiliations
- Muthén & Muthén (United States) (US)
- Venue
- Psychological Methods
- Published
- 2012-09
- Volume
- 17
- Issue
- 3
- Pages
- 313-335
- ISSN
- 1082-989X (ISSN-L)
- 1939-1463
- OpenAlex
- W2157335584
- Cited by
- 1,562
- References
- 43
- Citations by year
- 2026: 148 · 2025: 140 · 2024: 136 · 2023: 118 · 2022: 135 · 2021: 169 · 2020: 124 · 2019: 127 · 2018: 103 · 2017: 99 · 2016: 74 · 2015: 77 · 2014: 60 · 2013: 32 · 2012: 12
- FWCI
- 76.81
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Psychological Methods (publishedVersion)
- PubMed (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
- Bayesian structural equation modeling, confirmatory factor analysis, structural equation modeling, system identification, cross-loadings, residual correlation, model misspecification, posterior predictive checking, model modification, latent variable measurement, Big Five personality traits, science achievement
- MeSH terms
- Bayes Theorem, Factor Analysis, Statistical, Humans, Markov Chains, Monte Carlo Method, Neuropsychological Tests, Personality Tests, Models, Statistical, Likelihood Functions
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.