×

Alternating imputation posterior estimation of models with crossed random effects. (English) Zbl 1247.62188

Summary: Generalized linear mixed models or latent variable models for categorical data are difficult to estimate if the random effects or latent variables vary at non-nested levels, such as persons and test items. D.G. Clayton and J. Rasbash [J. R. Stat. Soc., Ser. A 162, 425–436 (1999)] suggested an Alternating Imputation Posterior (AIP) algorithm for approximate maximum likelihood estimation. For item response models with random item effects, the algorithm iterates between an item wing in which the item mean and variance are estimated for given person effects and a person wing in which the person mean and variance are estimated for given item effects. The person effects used for the item wing are sampled from the conditional posterior distribution estimated in the person wing and vice versa. Clayton and Rasbash used marginal quasi-likelihood (MQL) and penalized quasi-likelihood (PQL) estimation within the AIP algorithm, but this method has been shown to produce biased estimates in many situations, so we use maximum likelihood estimation with adaptive quadrature. We apply the proposed algorithm to the famous salamander mating data, comparing the estimates with many other methods, and to an educational testing data set. We also present a simulation study to assess performance of the AIP algorithm and the Laplace approximation with different numbers of items and persons and a range of item and person variances.

MSC:

62J12 Generalized linear models (logistic models)
62P15 Applications of statistics to psychology
65C60 Computational problems in statistics (MSC2010)

Software:

WinBUGS; Stata; MLwiN; lme4
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] Albers, W.; Does, R. J.M. M.; Imbos, T.; Janssen, M. P.E., A stochastic growth model applied to repeated test of academic knowledge, Psychometrika, 54, 451-466 (1989)
[2] Albert, P. S.; Follmann, D. A., Modeling repeated count data subject to informative dropout, Biometrics, 56, 667-677 (2000) · Zbl 1060.62573
[3] Baltagi, B. H., Econometric Analysis of Panel Data (2001), Wiley: Wiley Chichester
[4] Bates, D., Maechler, M., Dai, B., 2008. The lme4: linear mixed-effects models using S4 classes. URL: http://lme4.r-forge.r-project.org/; Bates, D., Maechler, M., Dai, B., 2008. The lme4: linear mixed-effects models using S4 classes. URL: http://lme4.r-forge.r-project.org/
[5] Bellio, R.; Varin, C., A pairwise likelihood approach to generalized linear models with crossed random effects, Statistical Modelling, 5, 217-227 (2005) · Zbl 1111.62058
[6] Bock, R. D.; Lieberman, M., Fitting a response model for \(n\) dichotomously scored items, Psychometrika, 33, 179-197 (1970)
[7] Bock, R. D.; Schilling, S. G., High-dimensional full-information item factor analysis, (Berkane, M., Latent Variable Modelling and Applications to Causality (1997), Springer: Springer New York), 164-176 · Zbl 0919.62058
[8] Booth, J. G.; Hobert, J. P., Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm, Journal of the Royal Statistical Society, Series B, 61, 265-285 (1999) · Zbl 0917.62058
[9] Breslow, N. E., Whither PQL?, (Lin, D. Y.; Heagerty, P. J., Proceedings of the Second Seattle Symposium in Biostatistics: Analysis of Correlated Data (2004), Springer: Springer New York), 1-22 · Zbl 1390.62128
[10] Breslow, N. E.; Clayton, D. G., Approximate inference in generalized linear mixed models, Journal of the American Statistical Association, 88, 9-25 (1993) · Zbl 0775.62195
[11] Breslow, N. E.; Lin, X., Bias correction in generalised linear mixed models with a single component of dispersion, Biometrika, 82, 81-91 (1995) · Zbl 0823.62059
[12] Briggs, D. C.; Wilson, M., Generalizability in item response modeling, Journal of Educational Measurement, 44, 131-155 (2007)
[13] Brooks, S. P.; Gelman, A., General methods for monitoring convergence of iterative simulations, Journal of Computational and Graphical Statistics, 7, 434-455 (1998)
[14] Browne, W. J.; Draper, D., A comparison of Bayesian and likelihood methods for fitting multilevel models, Bayesian Analysis, 1, 473-514 (2006) · Zbl 1331.62125
[15] Browne, W. J.; Goldstein, H.; Rasbash, J., Multiple membership multiple classification (MMMC) models, Statistical Modelling, 1, 103-124 (2001)
[16] Butler, J. S.; Moffitt, R., A computationally efficient quadrature procedure for the one-factor multinomial probit model, Econometrica, 50, 761-764 (1982) · Zbl 0483.62099
[17] Chaimongkol, S.; Huffer, F. W.; Kamata, A., A Bayesian approach for fitting a random effect differential item functioning across group units, Thailand Statistician, 4, 27-41 (2006) · Zbl 1110.62158
[18] Chang, H.; Stout, W., The asymptotic posterior normality of the latent trait in an IRT model, Psychometrika, 58, 37-52 (1993) · Zbl 0785.62099
[19] Clayton, D. G.; Rasbash, J., Estimation in large crossed random-effect models by data augmentation, Journal of the Royal Statistical Society, Series A, 162, 425-436 (1999)
[20] De Boeck, P., Random item IRT models, Psychometrika, 73, 533-559 (2008) · Zbl 1284.62699
[21] De Boeck, P.; Wilson, M., Explanatory Item Response Models: a Generalized Linear and Nonlinear Approach (2004), Springer: Springer New York · Zbl 1098.91002
[22] De Jong, M. G.; Steenkamp, J., Relaxing measurement invariance in cross-national consumer research using a hierarchical IRT model, Journal of Consumer Research, 34, 260-278 (2007)
[23] De Jong, M. G.; Steenkamp, J., Finite mixture multilevel multidimensional ordinal IRT models for large-scale cross-cultural research, Psychometrika, 75, 3-32 (2010) · Zbl 1234.62160
[24] Diaz, R. E., Comparison of PQL and Laplace 6 estimates of hierarchical linear models when comparing groups of small incident rates in cluster randomised trials, Computational Statistics and Data Analysis, 51, 2871-2888 (2007) · Zbl 1161.62386
[25] Doolaard, S., 1999. Schools in Change or School in Chain. University of Twente, The Netherlands.; Doolaard, S., 1999. Schools in Change or School in Chain. University of Twente, The Netherlands.
[26] Fox, J. P.; Glas, C. A.W., Bayesian estimation of a multilevel IRT model using Gibbs sampling, Psychometrika, 66, 269-286 (2001) · Zbl 1293.62242
[27] Gelfand, A.; Smith, A., Sampling based approaches to calculating marginal densities, Journal of the American Statistical Association, 90, 398-409 (1990) · Zbl 0702.62020
[28] Gelman, A., Prior distributions for variance parameters in hierarchical models, Bayesian Analysis, 1, 515-533 (2006) · Zbl 1331.62139
[29] Gelman, A.; Rubin, D. B., Inference from iterative simulation using multiple sequences, Statistical Sceience, 7, 457-472 (1992) · Zbl 1386.65060
[30] Glas, C. A.W.; van der Linden, W. J., Computerized adaptive testing with item cloning, Applied Psychological Measurement, 27, 247-261 (2003)
[31] Goldstein, H., Multilevel covariance component models, Biometrika, 74, 430-431 (1987) · Zbl 0619.62064
[32] Goldstein, H., Nonlinear multilevel models, with an application to discrete response data, Biometrika, 78, 45-51 (1991)
[33] Goldstein, H., Multilevel Statistical Models (2003), Arnold: Arnold London · Zbl 1014.62126
[34] Goldstein, H.; Rasbash, J., Improved approximations for multilevel models with binary responses, Journal of the Royal Statistical Society, Series A, 159, 505-513 (1996) · Zbl 1001.62518
[35] Goldstein, H., Rasbash, J., Plewis, I., Draper, D., Browne, W.J., Yang, M., Woodhouse, G., Healy, M., 1998. A User’s Guide to MLwiN. Multilevel Models Project, Institute of Education, University of London, London.; Goldstein, H., Rasbash, J., Plewis, I., Draper, D., Browne, W.J., Yang, M., Woodhouse, G., Healy, M., 1998. A User’s Guide to MLwiN. Multilevel Models Project, Institute of Education, University of London, London.
[36] Hedeker, D.; Gibbons, R. D., A random-effects ordinal regression model for multilevel analysis, Biometrics, 50, 933-944 (1994) · Zbl 0826.62049
[37] Janssen, R.; Schepers, J.; Peres, D., Models with item and item group predictors, (De Boeck, P.; Wilson, M., Explanatory Item Response Models (2004), Springer: Springer New York), 198-212
[38] Janssen, R.; Tuerlinckx, F.; Meulders, M.; De Boeck, P., A hierarchical IRT model for criterion-referenced measurement, Journal of Educational and Behavioral Statistics, 25, 285-306 (2000)
[39] Joe, H., Accuracy of Laplace approximation for discrete response mixed models, Computational Statistics and Data Analysis, 52, 5066-5074 (2008) · Zbl 1452.62537
[40] Johnson, M. S.; Sinharay, S., Calibration of polytomous item families using Bayesian hierarchical modeling, Applied Psychological Measurement, 29, 369-400 (2005)
[41] Karim, M. R.; Zeger, S. L., Generalized linear models with random effects: Salamander mating revisited, Biometrics, 48, 631-644 (1992)
[42] Lambert, P. C., Comment on article by Browne and Draper, Bayesian Analysis, 1, 543-546 (2006) · Zbl 1331.62151
[43] Lee, Y.; Nelder, J. A., Hierarchical generalized linear models (with discussion), Journal of the Royal Statistical Society, Series B, 58, 619-678 (1996) · Zbl 0880.62076
[44] Lee, Y.; Nelder, J. A., Double-hierarchical generalized linear models (with discussion), Journal of the Royal Statistical Society, Series C, 55, 1-29 (2006)
[45] Lee, Y.; Nelder, J. A.; Pawitan, Y., Generalized Linear Models With Random Effects: Unified Analysis Via H-Likelihood (2006), Chapman & Hall/CRC: Chapman & Hall/CRC Boca Raton, FL · Zbl 1110.62092
[46] Lesaffre, E.; Spiessens, B., On the effect of the number of quadrature points in a logistic random-effects model: an example, Journal of the Royal Statistical Society, Series C, 50, 325-335 (2001) · Zbl 1112.62307
[47] Lin, X.; Breslow, N. E., Bias correction in generalized linear mixed models with multiple components of dispersion, Journal of the American Statistical Association, 91, 1007-1016 (1996) · Zbl 0882.62059
[48] Liu, Q.; Pierce, D. A., A note on Gauss-Hermite quadrature, Biometrika, 81, 624-629 (1994) · Zbl 0813.65053
[49] McCaffrey, D. F.; Lockwood, J. R.; Koretz, D.; Louis, T. A., Models for value-added modeling of teacher effects, Journal of Educational and Behavioral Statistics, 29, 67-101 (2004)
[50] McCulloch, C. E., Maximum likelihood variance components estimation for binary data, Journal of the American Statistical Association, 89, 330-335 (1994) · Zbl 0800.62139
[51] McGilchrist, C. A., Estimation in generalized mixed models, Journal of the Royal Statistical Society, Series B, 56, 61-69 (1994) · Zbl 0800.62433
[52] Mislevy, R. J., Randomization-based inference about latent variables from complex samples, Psychometrika, 56, 177-196 (1991) · Zbl 0850.62854
[53] Mislevy, R. J.; Beaton, A. E.; Kaplan, B.; Sheehan, K. M., Estimating population characteristic from sparse matrix samples of item responses, Journal of Educational Measurement, 29, 133-161 (1992)
[54] Natarajan, R.; Kass, R. E., Reference Bayesian methods for generalized linear mixed model, Journal of the American Statistical Association, 95, 227-237 (2000) · Zbl 0996.62021
[55] Naylor, J. C.; Smith, A. F.M., Applications of a method for the efficient computation of posterior distributions, Journal of the Royal Statistical Society, Series C, 31, 214-225 (1982) · Zbl 0521.65017
[56] Neyman, J.; Scott, E. L., Consistent estimates based on partially consistent observation, Econometrica, 16, 1-32 (1948) · Zbl 0034.07602
[57] Noh, M.; Lee, Y., REML estimation for binary data in GLMMs, Journal of Multivariate Analysis, 98, 896-915 (2007) · Zbl 1113.62087
[58] Patterson, H. D.; Thompson, R., Recovery of inter-block information when block sizes are unequal, Biometrika, 58, 545-554 (1971) · Zbl 0228.62046
[59] Pinheiro, J. C.; Bates, D. M., Approximation to the log-likelihood function in the nonlinear mixed-effects model, Journal of Computational and Graphics and Statistics, 4, 12-35 (1995)
[60] Pinheiro, J. C.; Chao, E. C., Efficient Laplacian and adaptive Gaussian quadrature algorithms for multilevel generalized linear mixed models, Journal of Computational and Graphical Statistics, 15, 58-81 (2006)
[61] Rabe-Hesketh, S.; Skrondal, A., Multilevel and Longitudinal Modeling Using Stata (2008), Stata Press: Stata Press College Station, TX · Zbl 1274.62031
[62] Rabe-Hesketh, S.; Skrondal, A.; Pickles, A., Maximum likelihood estimation of limited and discrete dependent variable models with nested random effects, Journal of Econometrics, 128, 301-323 (2005) · Zbl 1336.62079
[63] Rasbash, J.; Browne, W. J., Non-hierarchical multilevel models, (de Leeuw, J.; Meijer, E., Handbook of Multilevel Analysis (2007), Springer: Springer New York), 333-336
[64] Rasbash, J.; Goldstein, H., Efficient analysis of mixed hierarchical and crossed random structures using a multilevel model, Journal of Educational Statistics, 337-350 (1994)
[65] Raudenbush, S. W., A crossed random effects model for unbalanced data with applications in cross-sectional and longitudinal research, Journal of Educational and Behavioral Statistics, 18, 321-349 (1993)
[66] Raudenbush, S. W.; Bryk, A. S., Hierarchical Linear Models (2002), Sage: Sage Thousand Oaks, CA
[67] Raudenbush, S. W.; Yang, M.; Yosef, M., Maximum likelihood for generalized linear models with nested random effects via high-order, multivariate Laplace approximation, Journal of Computational and Graphical Statistics, 9, 141-157 (2000)
[68] Rodríguez, G.; Goldman, N., An assessment of estimation procedures for multilevel models with binary responses, Journal of the Royal Statistical Society, Series A, 158, 73-89 (1995)
[69] Rodríguez, G.; Goldman, N., Improved estimation procedures for multilevel models with binary response: a case study, Journal of the Royal Statistical Society, Series A, 164, 339-355 (2001) · Zbl 1002.62507
[70] Rubin, D. B., Multiple Imputation for Nonresponse in Surveys (1987), Wiley: Wiley New York · Zbl 1070.62007
[71] Schall, R., Estimation in generalized linear-models with random effects, Biometrika, 78, 719-727 (1991) · Zbl 0850.62561
[72] Schilling, S.; Bock, R. D., High dimensional maximum marginal likelihood item factor analysis by adaptive quadrature, Psychometrika, 70, 533-555 (2005) · Zbl 1306.62497
[73] Sinharay, S.; Johnson, M. S.; Williamson, D. M., Calibrating item families and summerizing the results using family expected response functions, Journal of Educational and Behavioral Statistics, 28, 295-313 (2003)
[74] Spiegelhalter, D., Thomas, A., Best, N., 2003. WinBUGS version 1.4 [Computer program]. UK: MRC Biostatistics Unit, Institute of Public Health, Robinson Way, Cambridge CB2 2SR.; Spiegelhalter, D., Thomas, A., Best, N., 2003. WinBUGS version 1.4 [Computer program]. UK: MRC Biostatistics Unit, Institute of Public Health, Robinson Way, Cambridge CB2 2SR.
[75] StataCorp,, 2007. Statistical Software: Release 10.0. [Computer program]. College Station, TX, Stata Corporation.; StataCorp,, 2007. Statistical Software: Release 10.0. [Computer program]. College Station, TX, Stata Corporation.
[76] Stiratelli, R.; Laird, N. M.; Ware, J. H., Random effects models for serial observations with binary responses, Biometrics, 40, 961-971 (1984)
[77] Tanner, M. A., Tools for Statistical Inference: Methods for the Exploration of Posterior Distributions and Likelihood Functions (1996), Springer: Springer New York · Zbl 0846.62001
[78] Tanner, M. A.; Wong, W. H., The calculation of posterior distributions by data augmentation, Journal of the American Statistical Association, 82, 528-540 (1987) · Zbl 0619.62029
[79] Thomas, N.; Gan, N., Generating multiple imputations for matrix sampling data analyzed with item response models, Journal of Educational and Behavioral Statistics, 22, 425-445 (1997)
[80] Tibaldi, F. S.; Verbeke, G.; Molenberghs, G.; Renard, D.; Van den Noortgate, W.; De Boeck, P., Conditional mixed models with crossed random effects, British Journal of Mathematical and Statistical Psychology, 60, 351-365 (2007)
[81] Tierney, L.; Kadane, J. B., Accurate approximations for posterior moments and marginal densities, Journal of the American Statistical Association, 81, 82-86 (1986) · Zbl 0587.62067
[82] Vaida, F.; Meng, X. L., Two slice-EM algorithms for fitting generalized linear mixed models with binary response, Statistical Modelling, 5, 229-242 (2005) · Zbl 1111.62061
[83] Van den Noortgate, W.; De Boeck, P.; Meulders, M., Cross-classification multilevel logistic models in psychometrics, Journal of Educational and Behavioral Statistics, 28, 369-386 (2003)
[84] Vermunt, J., Multilevel mixture item response theory models: an application in education testing, Bulletin of the International Statistical Institute, 1-4 (2007)
[85] Wei, G. C.G.; Tanner, M. A., A Monte Carlo implementation of the EM algorithm and the poor man’s data augmentation algorithms, Journal of the American Statistical Association, 85, 699-704 (1990)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.