Effects of Within-Group Homogeneity on Parameter Estimation of the Multilevel Rasch Model

다국어 초록 (Multilingual Abstract)

If a hierarchical structure exists in educational measurement data and examinees within groups are homogeneous, a multilevel item response theory (MLIRT) model may be appropriate. Among the MLIRT models, the multilevel Rasch model is equivalent to a generalized linear mixed model (GLMM) with a logit link where person abilities are considered random effects and item difficulties fixed effects. Then the lme4 package in R can be used to fit the multilevel Rasch model. In this study, it was shown how the multilevel Rasch model can be formulated as a three-level GLMM, followed by a simulation analysis, where intraclass correlation (ICC) of latent abilities as a measure of within-group homogeneity was manipulated from low to high under the conditions of small to large numbers of examinees and items. Item parameter estimates by marginal maximum likelihood estimation (MMLE) were compared with those obtained under the GLMM framework. Biases of item parameter estimates by both methods were not evident in all conditions. However, estimation results by MMLE became proportionally less accurate as the ICC increased when the number of examinees was small. If the number of examinees was large, estimation accuracies of MMLE were acceptable even when a high level of within-group homogeneity existed. GLMM produced stable results at all levels of ICC. In all conditions, the number of examinees was more influential than the number of items.

참고문헌 (Reference)

1 Bates, D., "lme4: Linear mixed-effects models using S4 classes. R package (Version 0.999375-36) [Computer program]"

2 Spiegelhalter, D. J., "WinBUGS (Version 1.4) [Computer program]"

3 de Ayala, R. J., "The theory and practice of item response theory" The Guilford Press 2009

4 Adams, R. J., "The multidimensional random coefficients multinomial logit model" 21 : 1-23, 1997

5 De Boeck, P., "The estimation of item response models with the lmer function from the lme4 package in R" 39 : 1-28, 2011

6 R Development Core Team, "R: A language and environment for statistical computing"

7 Kamata, A., "Procedures to perform item response data analysis by HLM" 2002

8 Rasch, G., "Probabilistic models for some intelligence and attainment tests" Danish Institute for Educational Research 1960

9 Adams, R. J., "Multilevel item response models : An approach to errors in variable regression" 22 : 47-76, 1997

10 Snijders, T., "Multilevel analysis: An introduction to basic and advanced multilevel modeling" Sage 1999

11 Kamata, A., "Multilevel Rasch models, In Multivariate and mixture distribution Rasch models: Extensions and applications" Springer 2007

12 Fox, J. P., "Multilevel IRT using dichotomous and polytomous response data" 58 : 145-172, 2005

13 Fox, J. P., "Multilevel IRT modeling in practice with the package mlirt" 20 : 1-16, 2007

14 Wright, B. D., "Misunderstanding the Rasch model" 14 : 219-226, 1977

15 Bock, R. D., "Marginal maximum likelihood estimation of item parameters : Application of an EM algorithm" 46 : 443-459, 1981

16 Baker, F. B., "Item response theory: Parameter estimation techniques" Dekker 2004

17 Kamata, A., "Item analysis by the hierarchical generalized linear model" 38 : 79-93, 2001

18 Raudenbush, S. W., "Hierarchical linear models: Applications and data analysis methods" Sage 2002

19 Raudenbush, S. W., "HLM: Hierarchical Linear and Nonlinear Modeling [Computer program]"

20 McCulloch, C. E., "Generalized, linear, and mixed models" Wiley 2001

21 Doran, H., "Estimating the multilevel Rasch model : With the lme4 package" 20 : 1-18, 2007

22 Schwarz, G., "Estimating the dimension of a model" 6 : 461-464, 1978

23 Wu, M. L., "ConQuest: Multi-Aspect Test Software"

24 Albert, J. H., "Bayesian estimation of normal ogive item response curves using Gibbs sampling" 17 : 251-269, 1992

25 Natesan, P., "Bayesian estimation of graded response multilevel models using Gibbs sampling : Formulation and illustration" 70 : 420-439, 2010

26 Fox, J. P., "Bayesian estimation of a multilevel IRT model using Gibbs sampling" 66 : 269-286, 2001

27 Zimowski, M., "BILOG-MG (Version 3.0) [Computer program]"

28 Patz, R. J., "Applications and extensions of MCMC in IRT: Multiple item types, missing data, and rated responses" 24 : 342-366, 1999

29 Linacre, J. M., "A user's guide to WINSTEPS and MINISTEP: Rasch-model computer programs"

30 Patz, R. J., "A straightforward approach to Markov chain Monte Carlo methods for item response models" 24 : 146-178, 1999

31 Rijmen, F., "A nonlinear mixed model framework for item response theory" 8 : 185-205, 2003

32 Akaike, H., "A new look at the statistical model identification" 19 : 716-723, 1974