, Volume 58, Issue 2, pp 315–330 | Cite as

A latent class approach to fitting the weighted Euclidean model, clascal

  • Suzanne Winsberg
  • Geert De Soete


A weighted Euclidean distance model for analyzing three-way proximity data is proposed that incorporates a latent class approach. In this latent class weighted Euclidean model, the contribution to the distance function between two stimuli is per dimension weighted identically by all subjects in the same latent class. This model removes the rotational invariance of the classical multidimensional scaling model retaining psychologically meaningful dimensions, and drastically reduces the number of parameters in the traditional INDSCAL model. The probability density function for the data of a subject is posited to be a finite mixture of spherical multivariate normal densities. The maximum likelihood function is optimized by means of an EM algorithm; a modified Fisher scoring method is used to update the parameters in the M-step. A model selection strategy is proposed and illustrated on both real and artificial data.

Key words

weighted Euclidean distance model INDSCAL latent class analysis mixture distribution model EM algorithm 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Aitkin, M., Anderson, D., & Hinde, J. (1981). Statistical modelling of data on teaching styles.Journal of the Royal Statistical Society, Series A, 144, 419–461.Google Scholar
  2. Akaike, H. (1977). On entropy maximization. In P. R. Krishnaiah (Ed.),Applications of statistics (pp. 27–41). Amsterdam: North-Holland.Google Scholar
  3. Bozdogan, H. (1987). Model selection and Akaike's information criterion (AIC): The general theory and its analytical extensions.Psychometrika, 52, 345–370.Google Scholar
  4. Bozdogan, H. (1992).Choosing the number of component clusters in the mixture-model using a new informational complexity criterion of the inverse-Fisher information matrix. Paper presented at the 16th Annual Meeting of the German Classification Society, Dortmund, Germany.Google Scholar
  5. Böckenholt, U., & Böckenholt, I. (1990). Modeling individual differences in unfolding preference data: A restricted latent class approach.Applied Psychological Measurement, 14, 257–269.Google Scholar
  6. Böckenholt, U., & Böckenholt, I. (1991). Constrained latent class analysis: Simultaneous classification and scaling of discrete choice data.Psychometrika, 56, 699–716.Google Scholar
  7. Carroll, J. D., & Chang, J. J. (1970). Analysis of individual differences in multidimensional scaling via an N-way generalization of Eckart-Young decomposition.Psychometrika, 35, 283–319.Google Scholar
  8. Carroll, J. D., & Winsberg, S. (1991).Fitting an extended INDSCAL model to three-way proximity data. Unpublished manuscript, Rutgers University, Newark.Google Scholar
  9. de Leeuw, J., & Meulman, J. (1986). A special jackknife for multidimensional scaling.Journal of Classification, 3, 97–112.Google Scholar
  10. Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977). Maximum likelihood estimation from incomplete data via the EM algorithm.Journal of the Royal Statistical Society, Series B, 39, 1–38.Google Scholar
  11. DeSarbo, W. S., Howard, D. J., Jedidi, K. (1991). MULTICLUS: A new method for simultaneously performing multidimensional scaling and cluster analysis.Psychometrika, 56, 121–136.Google Scholar
  12. De Soete, G. (1990). A latent class approach to modeling pairwise preferential choice data. In M. Schader & W. Gaul (Eds.),Knowledge, data and computer-assisted decisions (pp. 103–113). Berlin: Springer-Verlag.Google Scholar
  13. De Soete, G. (1993). Using latent class analysis in categorization research. In I. Van Mechelen, J. Hampton, R. Michalski, & P. Theuns (Eds.),Categories and concepts: Theoretical views and inductive data analysis (pp. 309–330). London: Academic Press.Google Scholar
  14. De Soete, G., & DeSarbo, W. S. (1991). A latent class probit model for analyzing pick any/N data.Journal of Classification, 8, 45–63.Google Scholar
  15. De Soete, G., & Heiser, W. J. (1992).A latent class unfolding model for analyzing single stimulus preference ratings. Unpublished manuscript, University of Ghent, Belgium.Google Scholar
  16. De Soete, G., & Winsberg, S. (1993). A Thurstonian pairwise choice model with univariate and multivariate spline transformations.Psychometrika, 58, 233–256.Google Scholar
  17. De Soete, G., & Winsberg, S. (in press).A latent class vector model for analyzing preference ratings. Journal of Classification.Google Scholar
  18. Formann, A. K. (1989). Constrained latent class models: Some further applications.British Journal of Mathematical and Statistical Psychology, 42, 37–54.Google Scholar
  19. Gower, J. C. (1966). Some distance properties of latent root and vector methods using multivariate analysis.Biometrika, 53, 325–338.Google Scholar
  20. Hope, A. C. (1968). A simplified Monte Carlo significance test procedure.Journal of the Royal Statistical Society, Series B, 30, 582–598.Google Scholar
  21. Lazarsfeld, P. F., & Henry, R. W. (1968).Latent structure analysis. New York: Houghton Mifflin.Google Scholar
  22. McLachlan, G. J., & Basford, K. E. (1988).Mixture models. New York: Marcel Dekker.Google Scholar
  23. Ramsay, J. O. (1977). Maximum likelihood estimation in multidimensional scaling.Psychometrika, 42, 241–266.Google Scholar
  24. Ramsay, J. O. (1982). Some statistical approaches to multidimensional scaling data.Journal of the Royal Statistical Society, Series A, 145, 285–312.Google Scholar
  25. Ramsay, J. O. (1991).MULTISCALE manual (Extended version). Montreal: McGill University.Google Scholar
  26. Schwarz, G. (1978). Estimating the dimensions of a model.Annals of Statistics, 6, 461–464.Google Scholar
  27. Torgerson, W. S. (1958).Theory and methods of scaling. New York: Wiley.Google Scholar
  28. Winsberg, S., & Carroll, J. D. (1989a). A quasi-nonmetric method for multidimensional scaling via an extended Euclidean model.Psychometrika, 54, 217–229.Google Scholar
  29. Winsberg, S., & Carroll, J. D. (1989b). A quasi-nonmetric method for multidimensional scaling of multiway data via a restricted case of an extended INDSCAL model. In R. Coppi & S. Bolasco (Eds.),Multiway data analysis (pp. 405–414). Amsterdam: North-Holland.Google Scholar
  30. Winsberg, S., & Ramsay, J. O. (1983). Monotone spline transformations for dimension reduction.Psychometrika, 48, 575–595.Google Scholar

Copyright information

© The Psychometric Society 1993

Authors and Affiliations

  • Suzanne Winsberg
    • 1
  • Geert De Soete
    • 2
  1. 1.IRCAMParisFrance
  2. 2.Department of Data AnalysisUniversity of GhentGhentBelgium

Personalised recommendations