, Volume 44, Issue 4, pp 443–460 | Cite as

Maximum likelihood estimation of the polychoric correlation coefficient

  • Ulf Olsson


The polychoric correlation is discussed as a generalization of the tetrachoric correlation coefficient to more than two classes. Two estimation methods are discussed: Maximum likelihood estimation, and what may be called “two-step maximum likelihood” estimation. For the latter method, the thresholds are estimated in the first step. For both methods, asymptotic covariance matrices for estimates are derived, and the methods are illustrated and compared with artificial and real data.

Key words

ordinal data polychoric correlation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Reference note

  1. Pearson, K. Mathematical contributions to the theory of evolution. XIII: On the theory of contingency and its relation to association and normal correlation.Drapers Company Research Memoirs, Biometrics Series, No. 1, 1904.Google Scholar


  1. Brown, M. B. & Bendetti, J. K. On the mean and variance of the tetrachoric correlation coefficient.Psychometrika, 1977,42, 347–355.CrossRefGoogle Scholar
  2. Froemel, E. C. A comparison of computer routines for the calculation of the tetrachoric correlation coefficient.Psychometrika, 1971,36, 165–174.CrossRefGoogle Scholar
  3. Hamdan, M. A. On the structure of the tetrachoric series.Biometrika, 1968,55, 261–262.Google Scholar
  4. Hamdan, M. A. The equivalence of tetrachoric and maximum likelihood estimates ofρ in 2 × 2 tables.Biometrika, 1970,57, 212–215.Google Scholar
  5. Hamdan, M. A. On the polychoric series method for estimatingρ in contingency tables.Psychometrika, 1971,36, 253–259.Google Scholar
  6. IMSL Library 1 (Ed. 5). Houston, Texas: International Mathematical and Statistical Libraries, 1975.Google Scholar
  7. Johnson, N. L. & Kotz, S.Distributions in statistics: Continuous multivariate distributions. New York: Wiley, 1972.Google Scholar
  8. Kirk, D. B. On the numerical approximation of the bivariate normal (tetrachoric) correlation coefficient.Psychometrika, 1973,38, 259–268.CrossRefGoogle Scholar
  9. Lancaster, H. O. & Hamdan, M. A. Estimation of the correlation coefficient in contingency tables with possibly nonmetrical characters.Psychometrika, 1964,29, 383–391.CrossRefGoogle Scholar
  10. Martinson, E. O. & Hamdan, M. A. Maximum Likelihood and some other asymptotically efficient estimators of correlation in two way contingency tables.Journal of Statistical Computation and Simulation, 1971,1, 45–54.Google Scholar
  11. Martinson, E. O. & Hamdan, M. A. Algorithm AS 87: Calculation of the polychoric estimate of correlation in contingency tables.Applied Statistics, 1975,24, 272–278.Google Scholar
  12. Olsson, U. On the robustness of factor analysis against crude classification of the observations.Multivariate Behavioral Research, 1979 (in press).Google Scholar
  13. Pearson, K. Mathematical contributions to the theory of evolution, VII: On the correlation of characters not quantitatively measurable.Philosophical Transactions of the Royal Society of London, Series A, 1901,195, 1–47.Google Scholar
  14. Pearson, K. & Pearson E. S. On polychoric coefficients of correlation.Biometrika, 1922,14, 127–156.Google Scholar
  15. Ritchie-Scott, A. The correlation coefficient of a polychoric table.Biometrika, 1918,12, 93–133.Google Scholar
  16. Tallis, G. M. The maximum likelihood estimation of correlation from contingency tables.Biometrics, 1962,18, 342–353.Google Scholar
  17. Wainer, H. & Thissen, D. Three steps towards robust regression.Psychometrika, 1976,41, 9–34.CrossRefGoogle Scholar

Copyright information

© The Psychometric Society 1979

Authors and Affiliations

  • Ulf Olsson
    • 1
  1. 1.University of UppsalaSweden

Personalised recommendations