, Volume 42, Issue 1, pp 7–67 | Cite as

Nonmetric individual differences multidimensional scaling: An alternating least squares method with optimal scaling features

  • Yoshio Takane
  • Forrest W. Young
  • Jan de Leeuw


A new procedure is discussed which fits either the weighted or simple Euclidian model to data that may (a) be defined at either the nominal, ordinal, interval or ratio levels of measurement; (b) have missing observations; (c) be symmetric or asymmetric; (d) be conditional or unconditional; (e) be replicated or unreplicated; and (f) be continuous or discrete. Various special cases of the procedure include the most commonly used individual differences multidimensional scaling models, the familiar nonmetric multidimensional scaling model, and several other previously undiscussed variants.

The procedure optimizes the fit of the model directly to the data (not to scalar products determined from the data) by an alternating least squares procedure which is convergent, very quick, and relatively free from local minimum problems.

The procedure is evaluated via both Monte Carlo and empirical data. It is found to be robust in the face of measurement error, capable of recovering the true underlying configuration in the Monte Carlo situation, and capable of obtaining structures equivalent to those obtained by other less general procedures in the empirical situation.

Key words

Euclidian model INDSCAL measurement similarities data analysis similarities data quantification successive block algorithm 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Reference Notes

  1. Bloxom, B.Individual differences in multidimensional scaling (Research Bulletin 68-45). Princeton, N. J.: Educational Testing Service, 1968.Google Scholar
  2. Carroll, J. D. & Chang, J. J.IDIOSCAL (Individual Differences in Orientation Scaling). Paper presented at the Spring meeting of the Psychometric Society, Princeton, N. J., April, 1972.Google Scholar
  3. Carroll, J. D. & Chang, J. J.Some methodological advances in INDSCAL. Paper presented at the Spring meeting of the Psychometric Society, Stanford, California, April, 1974.Google Scholar
  4. de Leeuw, J.Canonical discriminant analysis of relational data (Research Bulletin RB004-75). Leiden, The Netherlands: Datatheorie, University of Leiden, 1975.Google Scholar
  5. de Leeuw, J.An initial estimate for INDSCAL. Unpublished note, 1974.Google Scholar
  6. de Leeuw, J.The positive orthant method for nonmetric multidimensional scaling (Research Note RN 001-70). Leiden, TheN etherlands: Datatheorie, University of Leiden, 1970.Google Scholar
  7. de Leeuw, J. & Pruzansky,^S.A new computational method to fit the weighted Euclidean model (SUMSCAL). Unpublished notes, Bell Laboratories, 1975.Google Scholar
  8. Gill, P. E. & Murray, W.Two methods for the solution of linearly constrained and unconstrained optimization problems (NPL Report NAC 25). Teddington, England: National Physics Laboratory, November, 1972.Google Scholar
  9. Guttman, L.Smallest space analysis by the absolute value principle. Paper presented at the symposium on “Theory and practice of measurement” at the Nineteenth International Congress of Psychology, London, 1969.Google Scholar
  10. Harshman, R. A.Foundations of the PARAFAC procedure: Models and conditions for an explanatory multi-modal factor analysis (Working Papers in Phonetics No. 16). Los Angeles: University of California, 1970.Google Scholar
  11. Horst, P.The prediction of personal adjustment (Bulletin 48). New York: The Social Science Research Council, 1941.Google Scholar
  12. Jacobowitz, D.The acquisition of semantic structures. Unpublished doctoral dissertation, University of North Carolina, 1975.Google Scholar
  13. Jones, L. E. & Wadington, J.Sensitivity of INDSCAL to simulated individual differences in dimension usage patterns and judgmental error. Paper delivered at the Spring meeting of the Psychometric Society, Chicago, April, 1973.Google Scholar
  14. Kruskal, J. B., Young, F. W., & Seery, J. B.How to use KYST, a very flexible program to do multidimensional scaling and unfolding. Unpublished manuscript, Bell Laboratories, 1973.Google Scholar
  15. Obenchain, R.Squared distance scaling as an alternative to principal components analysis. Unpublished notes, Bell Laboratories, 1971.Google Scholar
  16. Roskam, E. E.Data theory and algorithms for nonmetric scaling (parts 1 and 2). Unpublished manuscript, Catholic University, Nijmegen, The Netherlands, 1969.Google Scholar
  17. Yates, A.Nonmetric individual-differences multidimensional scaling with balanced least squares monotone regression. Paper presented at the Spring meeting of Psychometric Society, Princeton, N. J., April, 1972.Google Scholar
  18. Young, F. W.Polynomial conjoint analysis: Some second order partial derivatives (L. L. Thurstone Psychometric Laboratory Report, No. 108). Chapel Hill, North Carolina: The L. L. Thurstone Psychometric Laboratory, July, 1972.Google Scholar


  1. Bloxom, B. An alternative method of fitting a model of individual differences in multidimenlonal scaling.Psychometrika, 1974,39, 365–367.Google Scholar
  2. Bôcher, M.Introduction to higher algebra. New York: MacMillan, 1907.Google Scholar
  3. Carroll, J. D. & Chang, J. J. Analysis of individual differences in multidimensional scaling via anN-way generalization of “Eckart-Young” decomposition.Psychometrika, 1970,35, 238–319.Google Scholar
  4. Coombs, C. H.A theory of data. New York: Wiley, 1964.Google Scholar
  5. de Leeuw, J.Canonical analysis of categorical data. Leiden, the Netherlands: University of Leiden, 1973.Google Scholar
  6. de Leeuw, J., Young, F. W. & Takane, Y. Additive structure in qualitative data: An alternating least squares method with opitmal scaling features.Psychometrika, 1976,41, 471–503.Google Scholar
  7. Eckart, C. & Young, G. The approximation of one matrix by another of lower rank.Psychometrika, 1936,3, 211–218.Google Scholar
  8. Ekman, G. Dimensions of color vision.Journal of Psychology, 1954,38, 467–474.Google Scholar
  9. Fisher, R. A.Statistical methods for research workers (10th ed.). Edinburgh: Oliver and Boyd, 1946.Google Scholar
  10. Funk, S., Horowitz, A., Lipshitz, R. & Young, F. W. The perceived structure of American ethnic groups: The use of multidimensional scaling in stereotype research.Sociometry, in press.Google Scholar
  11. Green, P. E. & Rao, V. R.Applied multidimensional scaling: A comparison of approaches and algorithms. New York: Holt, Rinehart and Winston, 1972.Google Scholar
  12. Guttman, L. A general nonmetric technique for finding the smallest coordinate space for a configuration of points.Psychometrika, 1968,33, 469–506.Google Scholar
  13. Hageman, L. A. & Prosching, T. A. Aspects of nonlinear block successive overrelaxation.SIAM Journal of Numerical Analysis, 1975,12, 316–335.Google Scholar
  14. Hayashi, C. Minimum dimension analysis.Behaviormetrika, 1974,1, 1–24.Google Scholar
  15. Horan, C. B. Multidimensional scaling: Combining observations when individuals have different perceptual structures.Psychometrika, 1969,34, 139–165.Google Scholar
  16. Johnson, R. M. Pairwise nonmetric multidimensional scaling.Psychometrika, 1973,38, 11–18.Google Scholar
  17. Johnson, S. C. Hierarchical clustering schemes.Psychometrika, 1967,32, 241–254.Google Scholar
  18. Jones, L. E. & Young, F. W. The structure of a social environment: A longitudinal individual differences scaling of an intact group.Journal of Personality and Social Psychology, 1972,24, 108–121.Google Scholar
  19. Jöreskog, K. A general method for analysis of covariance structures.Biometrika, 1970,57, 239–251.Google Scholar
  20. Kruakal, J. B. Nonmetric multidimensional scaling.Psychometrika, 1964,29, 1–27; 115–129.Google Scholar
  21. Kruskal, J. B. & Carroll, J. D. Geometric models and badness-of-fit functions. In P. R. Krishnaiah,Multivariate analysis (Vol. 2), New York: Academic Press, Inc., 1969.Google Scholar
  22. Lawson, C. L. & Hanson, R. J.Solving least squares problems. Englewood Cliffs, N. J.: Prentice-Hall, 1974.Google Scholar
  23. Levin, J. Three-mode factor analysis.Psychological Bulletin, 1965,64, 442–452.Google Scholar
  24. Levinsohn, J. R. & Young, F. W. Two special-purpose programs that perform nonmetric multidimensional scaling.Journal of Marketing Research, 1974,11, 315–316.Google Scholar
  25. Lingoes, J. C.The Guttman-Lingoes nonmetric program series. Ann Arbor, Michigan: Mathesis Press, 1973.Google Scholar
  26. Lingoes, J. C. & Roskam, E. E. A mathematical and empirical analysis of two multidimensional scaling algorithms.Psychometrika Monograph Supplement, 1973,38 (4, Pt. 2).Google Scholar
  27. McGee, V. C. Multidimensional scaling ofn sets of similarity measures: A nonmetric individual differences approach.Multivariate Behavioral Research, 1968,3, 233–248.Google Scholar
  28. Messick, S. J. & Abelson, R. P. The additive constant problem in multidimensional scaling.Psychometrika, 1956,21, 1–15.Google Scholar
  29. Miller, G. A. & Nicely, P. E. An analysis of perceptual confusions among some English consonants.Journal of the Acoustical Society of America, 1953,27, 338–352.Google Scholar
  30. Peterson, G. E. & Barney, H. L. Control methods used in a study of the vowels.Journal of the Acoustical Society of America, 1952,24, 175–184.Google Scholar
  31. Schönemann, P. H. An algebraic solution for a class of subjective metric models.Psychometrika, 1972,37, 441–451.Google Scholar
  32. Shepard, R. N. The analysis of proximities: Multidimensional scaling with an unknown distance function.I. Psychometrika, 1962,27, 125–140. (a)Google Scholar
  33. Shepard, R. N. The analysis of proximities: Multidimensional scaling with an unknown distance function. II.Psychometrika, 1962,27, 219–246. (b)Google Scholar
  34. Shepard, R. N. Stimulus and response generalization: Tests of a model relating generalization of distance in psychological space.Journal of Experimental Psychology, 1958,55, 509–523.Google Scholar
  35. Spence, I. A Monte Carlo evaluation of three nonmetric multidimensional scaling algorithms.Psychometrika, 1972,37, 461–486.Google Scholar
  36. Stevens, S. S. Mathematics, measurement, and psychophysics. In S. S. Stevens (Ed.),Handbook of experimental psychology. New York: Wiley, 1951.Google Scholar
  37. Stoer, J. On the numerical solution of constrained least-squares problems.SIAM Journal of Numerical Analysis, 1971,8, 382–411.Google Scholar
  38. Torgerson, W. S. Multidimensional scaling: I. Theory and method.Psychometrika, 1952,17, 401–419.Google Scholar
  39. Tucker, L. R. Relations between multidimensional scaling and three-mode factor analysis.Psychometrika, 1972,37, 3–27.Google Scholar
  40. Tucker, L. R. Some mathematical notes on three-mode factor analysis.Psychometrika, 1966,31, 279–311.Google Scholar
  41. Wilf, H. S. The numerical solution of polynomial equations. In A. Ralston & W. S. Wilf (Eds.),Mathematical methods of digital computers (Vol. 1). New York: Wiley, 1960.Google Scholar
  42. Wold, H. & Lyttkens, E. Nonlinear iterative partial least squares (NIPALS) estimation procedures.Bulletin ISI, 1969,43, 29–47.Google Scholar
  43. Yates, F. The analysis of replicated experiments when the field results are incomplete.The Empire Journal of Experimental Agriculture, 1933,1, 129–142.Google Scholar
  44. Young, F. W. A model for polynomial conjoint analysis algorithms. In R. N. Shepard, A. K. Romney, & S. Nerlove (Eds.),Multidimensional scaling (Vol. 1). New York: Seminar Press, 1972.Google Scholar
  45. Young, F. W. Nonmetric multidimensional scaling: Recovery of metric information.Psychometrika, 1970,35, 455–474.Google Scholar
  46. Young, F. W. POLYCON: A program for multidimensionally scaling one-, two-, or three-way data in additive, difference, or multiplicative spaces.Behavioral Science, 1973,18, 152–155.Google Scholar
  47. Young, F. W. Methods for describing ordinal data with cardinal models.Journal of Mathematical Psychology, 1975,12, 416–436. (a)Google Scholar
  48. Young, F. W. Scaling replicated conditional rank-order data.Sociological Methodology, 1975,12, 129–170. (b)Google Scholar
  49. Young, F. W., de Leeuw, J., & Takane, Y. Regression with qualitative and quantitative variables: An alternating least squares method with optimal scaling features.Psychometrika, 1976,41, 505–529.Google Scholar
  50. Young, F. W. & Torgerson, W. S. TORSCA, a Fortran-IV program for nommetric multidimensional scaling.Behavioral Science, 1968,13, 343–344.Google Scholar

Copyright information

© Psychometric Society 1977

Authors and Affiliations

  • Yoshio Takane
    • 2
  • Forrest W. Young
    • 2
  • Jan de Leeuw
    • 1
  1. 1.University of LeidenThe Netherlands
  2. 2.Psychometric LabUniversity of North CarolinaChapel Hill

Personalised recommendations