High-dimensional maximum marginal likelihood item factor analysis by adaptive quadrature

Psychometrika volume 70, pages533–555(2005)Cite this article

Abstract

Although the Bock–Aitkin likelihood-based estimation method for factor analysis of dichotomous item response data has important advantages over classical analysis of item tetrachoric correlations, a serious limitation of the method is its reliance on fixed-point Gauss-Hermite (G-H) quadrature in the solution of the likelihood equations and likelihood-ratio tests. When the number of latent dimensions is large, computational considerations require that the number of quadrature points per dimension be few. But with large numbers of items, the dispersion of the likelihood, given the response pattern, becomes so small that the likelihood cannot be accurately evaluated with the sparse fixed points in the latent space. In this paper, we demonstrate that substantial improvement in accuracy can be obtained by adapting the quadrature points to the location and dispersion of the likelihood surfaces corresponding to each distinct pattern in the data. In particular, we show that adaptive G-H quadrature, combined with mean and covariance adjustments at each iteration of an EM algorithm, produces an accurate fast-converging solution with as few as two points per dimension. Evaluations of this method with simulated data are shown to yield accurate recovery of the generating factor loadings for models of upto eight dimensions. Unlike an earlier application of adaptive Gibbs sampling to this problem by Meng and Schilling, the simulations also confirm the validity of the present method in calculating likelihood-ratio chi-square statistics for determining the number of factors required in the model. Finally, we apply the method to a sample of real data from a test of teacher qualifications.

This is a preview of subscription content, log in to check access.

References

  1. Ahrens J.H., Dieter U. (1979) Computer methods for sampling from the exponential and normal distributions. Communications of the Association for Computing Machinery 15:873–882

    Google Scholar 

  2. Ansari A., Jedidi K. (2000) Bayesian factor analysis for multilevel binary observations. Psychometrika 65(4):475–496

    Google Scholar 

  3. Bartholomew D.J., Knott M. (1999) Latent Variable Models and Factor Analysis. Oxford, New York

    Google Scholar 

  4. Bock R.D. (1975/1985) Multivariate Statistical Methods in Behavioral Research. McGraw-Hill, New York; 1985 reprint, Chicago: Scientific Software International

  5. Bock R.D., Lieberman M. (1970) Fitting a response model for dichotomously scored items. Psychometrika 35:179–197

    Google Scholar 

  6. Bock R.D., Aitkin M. (1981) Marginal maximum likelihood estimation of item parameters: application of an EM algorithm. Psychometrika 46:443–459

    Google Scholar 

  7. Bock R.D., Gibbons R.D., Muraki E. (1987) Full information item factor analysis. Applied Psychological Measurement 12(3):261–280

    Google Scholar 

  8. Bock R.D., Schilling S.G. (1997) High-dimensional full-information item factor analysis. In: Birkane M. (ed) Latent Variable Modeling and Applications to Causality. Springer, New-York, pp 163–176

    Google Scholar 

  9. Bock R.D., Gibbons R.D., Muraki E., Schilling S.G., Wilson D.T., Wood R. (1999) TESTFACT 3: Test Scoring, Item Statistics, and Full-information Item Factor Analysis. Scientific Software International, Chicago

    Google Scholar 

  10. Divgi D.R. (1979) Calculation of the tetrachoric correlation coefficient. Psychometrika 44:169–172

    Google Scholar 

  11. Ferguson G.A. (1941) The factorial interpretation of test difficulty. Psychometrika 6:323–329

    Google Scholar 

  12. Fox J.P., Glas C.A.W. (2001) Bayesian estimation of a multilevel IRT model using Gibbs sampling. Psychometrika 66(2):271–288

    Google Scholar 

  13. Guilford J.P. (1941) The difficulty of a test and its factor composition. Psychometrika 6:66–77

    Google Scholar 

  14. Haberman S.J. (1977) Log-linear models and frequency tables with small expected cell counts. Annals of Statistics 5:1148–1169

    Google Scholar 

  15. Harman H.H. (1987) Modern Factor Analysis. University of Chicago Press, Chicago

    Google Scholar 

  16. Hedeker D., Gibbons R.D. (1994) A random-effects ordinal regression model for multilevel analysis. Biometrics 50:933–944

    Google Scholar 

  17. Hill H.C., Schilling S.G., Ball D.L. (2004) Developing measures of teachers mathematics knowledge for teaching. Elementary School Journal, in press.

  18. Householder A.S. (1964) The Theory of Matrices in Numerical Analysis. Blaisdell, New York

    Google Scholar 

  19. Kaiser H.F. (1958) The varimax criterion for analytic rotation in factor analysis. Psychometrika 23:187–200

    Google Scholar 

  20. Leonelli B.T., Chang C.H., Bock R.D., Schilling S.G. (2000) A full-information item factor analysis interpretation of the MMPI-2: Normative Sampling with Non-pathonomic Descriptors. Journal of Personality Assessment 74(3):400–422

    Google Scholar 

  21. Lesaffre E., Spiessens B. (2001) On the effect of the number of quadrature points in a logistic random-effects model: an example. Applied Statistics 50:325–335

    Google Scholar 

  22. Lindstrom M.J., Bates D.M. (1990) Nonlinear mixed effects models for repeated measures data. Biometrics 46:673–687

    Google Scholar 

  23. Liu C., Rubin D.B., Wu Y.N. (1998) Parameter expansion to accelerate EM: The PX-EM algorithm. Biometrika 85(4):755–770

    Google Scholar 

  24. Liu Q., Pierce D.A. (1994) A note on G-H quadrature. Biometrika 81(3):624–629

    Google Scholar 

  25. Meng X.L., Schilling S.G. (1996) Fitting full-information factor models and an empirical investigation of bridge sampling. Journal of the American Statistical Association 91:1254–1267

    Google Scholar 

  26. Mislevy R.J. (1984) Estimating latent distributions. Psychometrika 49(3):359–381

    Google Scholar 

  27. Muthén B.O. (1984) A general structural equation model with dichotomous, ordered categorical and continuous latent variable indicators. Psychometrika 49:115–132

    Google Scholar 

  28. Muthén L. K., Muthén B.O. (1998–2001). Mplus User’s Guide (Second edition). Muthén & Muthén, Los Angeles CA

  29. Naylor J.C., Smith A.F.M. (1982) Applications of a method for the efficient computation of posterior distributions. Applied Statistics 31:214–225

    Google Scholar 

  30. Polak E. (1971) Computational Methods in Optimization. Academic Press, New York

    Google Scholar 

  31. Powell M.J.D. (1964) An efficient method for several variables without calculating derivatives. Computer Journal 7:155–162

    Google Scholar 

  32. Rabe-Hesketh S., Pickles A., Skrondal A., (2001) GLLAMM Manual. Tech. rept. 2001/01. Department of Biostatistics and Computing, Institute of Psychiatry, King’s College, University of London. Downloadable from http://www.gllamm.org.

  33. Rabe-Hesketh S., Skrondal A., Pickles A. (2002) Reliable estimation of generalized linear mixed models using adaptive quadrature. The Stata Journal 2:1–21

    Google Scholar 

  34. Rabe-Hesketh S., Skrondal A., Pickles A. (2005a) Generalized multilevel structural equation modeling. Psychometrika 69:167–190

    Google Scholar 

  35. Rabe-Hesketh S., Skrondal A., Pickles A. (2005b) Maximum likelihood estimation of limited and discrete dependent variable models with nested random effects. Journal of Econometrics, in press.

  36. Ramsay J.O. (1998) Estimating smooth monotone functions. Journal of the Royal Statistical Society, Series B. 60:365–375

    Google Scholar 

  37. Raudenbush S.W., Yang M., Yosef (2000) Maximum likelihood for generalized linear models with nested random effects via high-order, multivariate Laplace approximation. Journal of Computational and Graphical Statistics 9(1):141–157

    Google Scholar 

  38. Raudenbush S.W., Birk A.S. (2002) Maximum likelihood for generalized linear models with nested random effects via high-order, multivariate Laplace approximation. Journal of Computational and Graphical Statistics 9(1):141–157

    Google Scholar 

  39. Ripley B.D. (1987) Stochastic Simulation. Wiley, New York

    Google Scholar 

  40. Schilling S.G. (1993) Advances in Full Information Item Factor Analysis using the Gibbs Sampler. (Unpublished doctoral dissertation, University of Chicago)

  41. Schrage L. (1979) A more portable fortran random number generator. Association for Computing Machinery: Transactions on Mathematical Software 5:132–138

    Google Scholar 

  42. Thurstone L.L. (1947) Multiple Factor Analysis. The University of Chicago Press, Chicago

    Google Scholar 

  43. Thurstone L.L., Thurstone T.G. (1941) Factorial studies of intelligence. Psychometric Monographs No. 2. University of Chicago Press, Chicago

    Google Scholar 

  44. Tierney L., Kadane J.B. (1986) Accurate approximations for posterior moments and marginal densities. Journal of the American Statistical Association 81:82–86

    Google Scholar 

  45. Wei G.C.G., Tanner M.A. (1990) A Monte Carlo implementation of the EM algorithm and the poor man’s data augmentation algorithm. Journal of the American Statistical Association 85:699–704

    Google Scholar 

  46. Wood R., Wilson D.T., Gibbons R.D., Schilling S.G., Muraki E., Bock R.D. (2003) TESTFACF 4: Test Scoring, Item Statistics, and Full-information Item Factor Analysis. Scientific Software International, Chicago

    Google Scholar 

  47. Zimowski M.F., Muraki E., Mislevy R.J., Bock R.D. (1995) BILOG-MG: multiple-group item analysis and test scoring. Scientific Software International, Chicago

    Google Scholar 

Download references

Author information

Affiliations

  1. School of Education, University of Michigan, Ann Arbor, MI, 48109, USA

    Stephen Schilling

  2. Center for Health Statistics, University of Illinois at Chicago, USA

    R. Darrell. Bock

Authors
  1. Stephen Schilling
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. R. Darrell. Bock
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Stephen Schilling.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Schilling, S., Bock, R.D. High-dimensional maximum marginal likelihood item factor analysis by adaptive quadrature. Psychometrika 70, 533–555 (2005). https://doi.org/10.1007/s11336-003-1141-x

Download citation

Keywords

  • factor analysis
  • item response theory
  • latent variables
  • EM algorithm
  • marginal likelihood estimation
  • GLS estimation
  • adaptive quadrature
  • monte carlo integration