Hostname: page-component-5f745c7db-rgzdr Total loading time: 0 Render date: 2025-01-06T07:18:31.870Z Has data issue: true hasContentIssue false
  • English
  • Français

A Monte Carlo Evaluation of Three Nonmetric Multidimensional Scaling Algorithms

Published online by Cambridge University Press:  01 January 2025

Ian Spence
Ian Spence*
Affiliation:
University of Western Ontario
Get access Rights & Permissions [Opens in a new window]

Abstract

As a consequence of the complexity of the iterative optimum seeking procedures used by practical nonmetric multidimensional scaling algorithms, many of their computational properties have not been well understood. In particular, the following questions are of interest: (a) from the user’s point of view, are there significant differences between alternative available programs, (b) are suboptimal solutions frequently encountered, and how does this depend on the characteristics of the algorithm? Using Monte Carlo techniques to generate dissimilarity matrices with known underlying configurations, Kruskal’s M-D-SCAL, Guttman-Lingoes’ SSA-I, and Young-Torgerson’s TORSCA-9 programs were compared. It was found: (a) that differences between the solutions obtained by the algorithms were typically so small as to be of little practical importance, (b) deviant solutions were occasionally produced by each of the algorithms, but most often by M-D-SCAL, and furthermore, most frequently in one dimension. The likelihood of being trapped in a nonoptimal position is reduced by a good choice of initial configuration, and also possibly by constructing the iterative process to favor the possibility of stepping over small local depressions when far from the location of the optimum. The conclusions of this study are based on a total of 2160 scaling solutions.

Type
Original Paper
Information
Psychometrika , Volume 37 , Issue 4 , December 1972 , pp. 461 - 486
Copyright
Copyright © 1972 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

*

This paper is based on a study which formed part of a thesis submitted to the University of Toronto in conformity with the requirements for the degree of Doctor of Philosophy. I am indebted to a large number of people for advice and criticism; these include J. C. Ogilvie, C. D. Creelman, R. S. Lockhart, and S. Nishisato. I am also grateful for a number of useful criticisms and suggestions by the referees. This research was supported by a predoctoral fellowship and by grant APA-176 to J. C. Ogilvie from the National Research Council of Canada.

References

Box, M. J. A comparison of several current optimization methods, and the use of transformations in constrained problems. Computer Journal, 1966, 9, 67–77CrossRefGoogle Scholar
De Leeuw, J. The positive orthant method for nonmetric multidimensional scaling, 1969, The Netherlands: University of LeidenGoogle Scholar
Fletcher, R. A review of methods for unconstrained optimization. In Fletcher, R. (Eds.), Optimization, 1969, New York: Academic PressGoogle Scholar
Gleason, T. C. A general model for nonmetric multidimensional scaling, 1967, Ann Arbor, Michigan: University of MichiganGoogle Scholar
Gower, J. C. Some distance properties of latent root and vector methods used in multivariate analysis. Biometrika, 1966, 53, 325–338CrossRefGoogle Scholar
Guttman, L. A general nonmetric technique for finding the smallest coordinate space for a configuration of points. Psychometrika, 1968, 33, 469–506CrossRefGoogle Scholar
Kruskal, J. B. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika, 1964, 29, 1–27 (a)CrossRefGoogle Scholar
Kruskal, J. B. Nonmetric multidimensional scaling: A numerical method. Psychometrika, 1964, 29, 115–129 (b)CrossRefGoogle Scholar
Kruskal, J. B. How to use M-D-SCAL, a program to do multidimensional scaling and multidimensional unfolding, 1968, Murray Hill, New Jersey: Bell Telephone LaboratoriesGoogle Scholar
Lingoes, J. C. An IBM 7090 program for Guttman-Lingoes smallest space analysis—1. Behavioral Science, 1965, 10, 183–184Google Scholar
Lingoes, J. C., Roskam, E. E. A mathematical and empirical study of two multidimensional scaling algorithms, 1971, Ann Arbor, Michigan: University of MichiganGoogle Scholar
McGee, V. E. The multidimensional analysis of “elastic” distances. British Journal of Mathematical and Statistical Psychology, 1966, 19, 181–196CrossRefGoogle Scholar
McGee, V. E. A reply to some criticisms of elastic multidimensional scaling. British Journal of Mathematical and Statistical Psychology, 1967, 20, 243–247CrossRefGoogle ScholarPubMed
McGee, V. E. Speech perception and the development of appropriate nonmetric analytical techniques. Office of Naval Research Report. NONR 3897-05. Washington, D. C., July 1969.Google Scholar
Powell, M. J. D. A survey of numerical methods for unconstrained optimization. SIAM Review, 1970, 12, 79–97CrossRefGoogle Scholar
Ramsey, J. O. Some statistical considerations in multidimensional scaling. Psychometrika, 1969, 34, 167–182CrossRefGoogle Scholar
Roskam, E. E. Metric analysis of ordinal data in psychology, 1968, Voorschoten, The Netherlands: VAMGoogle Scholar
Roskam, E. E. A comparison of principles for algorithm construction in nonmetric scaling, 1969, Ann Arbor, Michigan: University of Michigan (a)Google Scholar
Roskam, E. E. Data theory and algorithms for nonmetric scaling, 1969, The Netherlands: University of Nijmegen (b)Google Scholar
Schönemann, P. H. Fitting a simplex symmetrically. Psychometrika, 1970, 35, 1–21CrossRefGoogle Scholar
Sherman, C. R. Nonmetric multidimensional scaling: A Monte Carlo study of the basic parameters. Psychometrika, 1972, 37, 323–355CrossRefGoogle Scholar
Spaeth, H. J., & Guthery, S. B. The use and utility of the monotone criterion in multidimensional scaling. Multivariate Behavioral Research, 1969, 4, 501–515CrossRefGoogle ScholarPubMed
Spang, H. A. A review of minimization techniques for nonlinear functions. SIAM Review, 1962, 4, 343–365CrossRefGoogle Scholar
Spence, I. Multidimensional scaling: An empirical and theoretical investigation. Unpublished Ph.D. Thesis, University of Toronto, 1970.Google Scholar
Torgerson, W. S. Theory and methods of scaling, 1958, New York: WileyGoogle Scholar
Whitley, V. M. Problems in multidimensional scaling. Unpublished Ph.D. Thesis, University of California at Irvine, 1971.Google Scholar
Wilde, D. J. Optimum seeking methods, 1964, Englewood Cliffs, New Jersey: Prentice HallGoogle Scholar
Young, F. W. A FORTRAN IV program for nonmetric multidimensional scaling, 1968, Chapel Hill, N. C.: University of North CarolinaGoogle Scholar
Young, F. W. Nonmetric multidimensional scaling: Recovery of metric information. Psychometrika, 1970, 35, 455–473CrossRefGoogle Scholar
Young, F. W., & Appelbaum, M. Nonmetric multidimensional scaling: The relationship of several methods, 1968, Chapel Hill, N. C.: University of North CarolinaGoogle Scholar
Young, G., & Householder, A. S. Discussion of a set of points in terms of their mutual distances. Psychometrika, 1938, 3, 19–22CrossRefGoogle Scholar