, Volume 46, Issue 4, pp 443–459 | Cite as

Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm

  • R. Darrell Bock
  • Murray Aitkin


Maximum likelihood estimation of item parameters in the marginal distribution, integrating over the distribution of ability, becomes practical when computing procedures based on an EM algorithm are used. By characterizing the ability distribution empirically, arbitrary assumptions about its form are avoided. The Em procedure is shown to apply to general item-response models lacking simple sufficient statistics for ability. This includes models with more than one latent dimension.

Key words

estimation of item parameters EM algorithm item analysis latent trait dichotomous factor analysis Law School Aptitude Test (LSAT) 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Reference notes

  1. Thissen, D. Personal communication, 1979.Google Scholar
  2. Thissen, D.Marginal maximum likelihood estimation for the one-parameter logistic model. Manuscript submitted for publication, 1981.Google Scholar


  1. Andersen, E. B. Conditional inference for multiple-choice questionnaires.British Journal of Mathematical and Statistical Psychology, 1973,26, 31–44.Google Scholar
  2. Andersen, E. B.Discrete statistical models with social science applications. Amsterdam: North-Holland, 1980.Google Scholar
  3. Andersen, E. B. & Madsen, M. Estimating parameters of the latent population distribution.Psychometrika, 1977,42, 357–374.Google Scholar
  4. Bartholomew, D. J. Factor analysis for categorical data (with Discussion).Journal of the Royal Statistical Society, Series B, 1980,42, 293–321.Google Scholar
  5. Bock, R. D. Estimating item parameters and latent ability when responses are scored in two or more nominal categories.Psychometrika, 1972,37, 29–51.Google Scholar
  6. Bock, R. D. Review ofThe dependability of behavioral measurements by Cronbach, L. J., Gleser, G. C., Nanda, H., & Rajaratnam, N.Science, 1972a,178, 1275A-1275A.Google Scholar
  7. Bock, R. D., & Jones, L. V.The measurement and prediction of judgment and choice. Chicago: International Educational Services, 1980 (reprint).Google Scholar
  8. Bock, R. D., & Lieberman, M. Fitting a response model forn dichotomously scored items.Psychometrika, 1970,35, 179–197.Google Scholar
  9. Christoffersson, A. Factor analysis of dichotomized variables.Psychometrika, 1975,40, 5–32.Google Scholar
  10. Dempster, A. P., Laird, N. M., & Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm (with Discussion).Journal of the Royal Statistical Society, Series B, 1977,39, 1–38.Google Scholar
  11. Fischer, G. Linear logistic test model as an instrument in educational research.Acta Psychologica, 1973,37, 359–374.Google Scholar
  12. Fischer, G. H. Individual testing on the basis of the dichotomous Rasch model. In L. J. T. Van der Kamp, W. F. Langerak, & D. M. N. de Gruijter (Eds).Psychometrics for Educational Debates. Chichester: Wiley, 171–188, 1980.Google Scholar
  13. Goldstein, H. Dimensionality, bias, independence, and measurement scale problems in latent trait test score models.British Journal of Mathematical and Statistical Psychology, 1980,33, 234–246.Google Scholar
  14. Haberman, S. J. Maximum likelihood estimates in exponential response models.Annals of Statistics, 1977,5, 815–841.Google Scholar
  15. Kaiser, H. F. The varimax criterion for analytic rotation in factor analysis.Psychometrika, 1958,23, 187–200.Google Scholar
  16. Kolakowski, D. & Bock, R. D.NORMOG: Maximum likelihood item analysis and test scoring; normal ogive model. Chicago: International Educational Services, 1973a.Google Scholar
  17. Kolakowski, D. & Bock, R. D.LOGOG: Maximum likelihood item analysis and test scoring; logistic model for multiple item responses. Chicago: International Educational Services, 1973b.Google Scholar
  18. Laird, N. M. Nonparametric maximum likelihood estimation of a mixing distribution.Journal of the American Statistical Association, 1978,73, 805–811.Google Scholar
  19. Lord, F. M. Practical applications of item characteristic curve theory.Journal of Educational Measurement, 1977,14, 117–138.Google Scholar
  20. Lord, F. M.Applications of item response theory to practical testing problems. New York: Erlbaum Associates, 1980.Google Scholar
  21. Lord, F. M., & Novick, M. R.Statistical theories of mental test scores. Reading (Mass.): Addison-Wesley, 1968.Google Scholar
  22. Micko, H. C. A psychological scale for reaction time measurement.Acta Psychologica, 1969,30, 324–335.Google Scholar
  23. Muthén, B. Contributions to factor analysis of dichotomized variables.Psychometrika, 1978,43, 551–560.Google Scholar
  24. Ramsay, J. O. Solving implicit equations in psychometric data analysis.Psychometrika, 1975,40, 337–360.Google Scholar
  25. Rao, C. R.Linear statistical inference and its applications (2nd ed.). New York: Wiley, 1973.Google Scholar
  26. Samejima, F. Estimation of latent ability using a response pattern of graded scores.Psychometric Monograph, No. 17, 1969.Google Scholar
  27. Sanathanan, L. & Blumenthal, S. The logistic model and estimation of latent structure.Journal of the American Statistical Association, 1978,73, 794–799.Google Scholar
  28. Scheiblechner, H. Das Lernen und Losen komplexer Denkaufgaben.Zeitschrift fur Experimentelle und Angewandte Psychologie, 1972,19, 476–506.Google Scholar
  29. Stroud, A. H., & Sechrest, D.Gaussian quadrature formulas. Englewood Cliffs (N.J.): Prentice-Hall, 1966.Google Scholar
  30. Thurstone, L. L.Multiple factor analysis. Chicago: University of Chicago Press, 1947.Google Scholar
  31. Wood, R. L. Wingersky & Lord, F. M.LOGIST: A computer program for estimating examinee ability and item characteristic curve parameters. Princeton (N.J.): Educational Testing Service, 1976.Google Scholar
  32. Wright, B. D. & Panchapakesan, N. A. A procedure for sample-free item analysis.Educational and Psychological Measurement, 1969,29, 23–48.Google Scholar

Copyright information

© The Psychometric Society 1981

Authors and Affiliations

  • R. Darrell Bock
    • 1
  • Murray Aitkin
    • 2
  1. 1.Department of Behavioral SciencesThe University of ChicagoChicago
  2. 2.University of LancasterUK

Personalised recommendations