Skip to main content
Log in

Hierarchical relations among three-way methods

  • Published:
Psychometrika Aims and scope Submit manuscript

Abstract

A number of methods for the analysis of three-way data are described and shown to be variants of principal components analysis (PCA) of the two-way supermatrix in which each two-way slice is “strung out” into a column vector. The methods are shown to form a hierarchy such that each method is a constrained variant of its predecessor. A strategy is suggested to determine which of the methods yields the most useful description of a given three-way data set.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Bentler, P. M. (1973). Assessment of developmental factor change at the individual and group level. In J. A. Nesselroade & H. W. Reese (Eds.),Life-span developmental psychology: Methodological issues (pp. 145–174). New York: Academic Press.

    Google Scholar 

  • Carroll, J. D. (1968). Generalization of canonical correlation analysis to three or more sets of variables.Proceedings of the 76th Convention of the American Psychological Association 3, 227–228.

    Google Scholar 

  • Carroll, J. D., & Arabie, P. (1980). Multidimensional scaling.Annual Review of Psychology, 31, 607–649.

    Google Scholar 

  • 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 

  • Carroll, J. D., & Wish, M. (1974). Models and methods for three-way multidimensional scaling. In D. H. Krantz, R. C. Atkinson, R. D. Luce, & P. Suppes (Eds.)Contemporary developments in mathematical psychology, Vol. II: Measurement, psychophysics, and neural information processing (pp. 57–105). San Francisco: Freeman & Co.

    Google Scholar 

  • Doledec, S., & Chessel, D. (1987) Rythmes saisonniers et composantes stationnelles en milieu aquatique. I.—Description d'un plan d'observation complet par projection de variables [Seasonal rhythms and stationary components in water environments. I.—Description of a complete observation scheme by projection of variables].Acta Oecologia/Oecologia Generalis, 8, 403–406.

    Google Scholar 

  • Escofier, B., & Pagès, J. (1983). Méthode pour l'analyse de plusieurs groupes de variables—Application a la caractérisation de vins rouges du Val de Loire [Method for the analysis of several groups of variables—Application to the characterization of red wines from the Loire valley].Revue de Statistique Appliquée, 31, 43–59.

    Google Scholar 

  • Escofier, B., & Pagès, J. (1984).L'analyse factorielle multiple [Multiple factor analysis]. Cahiers du bureau universitaire de recherche opérationnelle, no. 42, Université Pierre et Marie Curie, Paris.

    Google Scholar 

  • Escoufier, Y. (1973). Le traitement des variables vectorielles [The treatment of vector-valued variables].Biometrics, 29, 751–760.

    Google Scholar 

  • Glaçon, F. (1981).Analyse conjointe de plusieurs matrices de données [Simultaneous analysis of several data matrices]. Unpublished doctoral dissertation, University of Grenoble, France.

    Google Scholar 

  • Gower, J. C. (1966). Some distance properties of latent root and vector methods used in multivariate analysis.Biometrika, 53, 325–338.

    Google Scholar 

  • Harshman, R. A. (1970). Foundations of the PARAFAC procedure: models and conditions for an “explanatory” multi-mode factor analysis.UCLA Working Papers in Phonetics, 16, 1–84.

    Google Scholar 

  • Harshman, R. A. (1972). PARAFAC2: Mathematical and technical notes.UCLA Working Papers in Phonetics, 22, 31–44.

    Google Scholar 

  • Harshman, R. A., & Lundy, M. E. (1984a). Data preprocessing and the extended PARAFAC model. In H. G. Law, C. W. Snyder, J. A. Hattie, & R. P. McDonald (Eds.),Research methods for multimode data analysis (pp. 216–284). New York: Praeger.

    Google Scholar 

  • Harshman, R. A., & Lundy, M. E. (1984b). The PARAFAC model for three-way factor analysis and multidimensional scaling. In H. G. Law, C. W. Synder, J. A. Hattie, & R. P. McDonald (Eds.),Research methods for multimode data analysis (pp. 122–215). New York: Praeger.

    Google Scholar 

  • Jaffrennou, P. A. (1978).Sur l'analyse des familles finies de variables vectorielles [On the analysis of finite families of vector-valued variables]. University of Saint-Étienne, France.

    Google Scholar 

  • Kroonenberg, P. M. (1983).Three mode principal component analysis: Theory and applications. Leiden: DSWO press.

    Google Scholar 

  • Kroonenberg, P. M. (1989). The analysis of multiple tables in factorial ecology: III Three-mode principal component analysis: “Analyse triadique complète”.Acta Oecologia/Oecologia Generalis, 10, 245–256.

    Google Scholar 

  • Kroonenberg, P. M., & de Leeuw, J. (1980). Principal component analysis of three-mode data by means of alternating least squares algorithms.Psychometrika, 45, 69–97.

    Google Scholar 

  • Lavit, C. (1985). Application de la méthode STATIS [Application of the STATIS method].Statistique et Analyse des Données, 10, 103–116.

    Google Scholar 

  • Lavit, C. (1988).Analyse conjointe de tableaux quantitatifs [Simultaneous analysis of quantitative data sets]. Paris: Masson.

    Google Scholar 

  • Lechevallier, F. (1987).L'Analyse de l'evolution dans STATIS: Une solution et des généralisations [The analysis of evolution in STATIS: A solution and some generalizations]. Document de travail, UFR Sciences Economiques et Sociales, Villeneuve-d'Ascq, France.

    Google Scholar 

  • Levin, J. (1966). Simultaneous factor analysis of several gramian matrices.Psychometrika, 31, 413–419.

    Google Scholar 

  • L'Hermier des Plantes, H. (1976).Structuration des tableaux à trois indices de la statistique [Structuring statistical three-way data matrices]. Unpublished doctoral dissertation, University of Montpellier, France.

    Google Scholar 

  • Lundy, M. E., Harshman, R. A., & Kruskal, J. B. (1989). A two-stage procedure incorporating good features of both trilinear and quadrilinear models. In R. Coppi & S. Bolasco (Eds.),Multiway data analysis (pp. 123–130). Amsterdam: Elsevier Science Publishers.

    Google Scholar 

  • Pontier, J., Pernin, M. O., & Pagès, M. (1985, May).LONGI: Une méthode d'analyse de données longitudinales multivariées [LONGI: A method for the analysis of longitudinal multivariate data]. Paper presented at the Journées de Statistique, Pau, France.

  • Sabatier, R. (1987).Méthodes factorielles en analyse des données: Approximations et prise en compte de variables concomitantes [Factorial methods in data analysis: Approximations and accounting for instrumental variables]. Unpublished doctoral dissertation, University of Montpellier, France.

    Google Scholar 

  • ten Berge, J. M. F., & Kiers, H. A. L. (1991). Some clarifications of the CANDECOMP algorithm applied to INDSCAL.Psychometrika, 56, 317–326.

    Google Scholar 

  • Thioulouse, J., & Chessel, D. (1987). Les analyses multitableaus en écologie factorielle. I.—De la typologie d'état à la typologie de fonctionnement par l'analyse triadique [The analysis of multiple tables in factorial ecology. I.—From state typology to function typology by means of triadique analysis].Acta Oecologia/Oecologia Generalis, 8, 463–480.

    Google Scholar 

  • Tucker, L. R. (1966). Some mathematical notes on three-mode factor analysis.Psychometrika, 31, 279–311.

    Google Scholar 

  • Tucker, L. R. (1972). Relations between multidimensional scaling and three-mode factor analysis.Psychometrika, 37, 3–27.

    Google Scholar 

  • Tucker, L. R., & Messick, S. (1963). An individual differences model for multidimensional scaling.Psychometrika, 28, 333–367.

    Google Scholar 

  • Weesie, J., & Van Houwelingen, H. (1983).GEPCAM Users' manual: Generalized principal components analysis with missing values. Institute of Mathematical Statistics, University of Utrecht, The Netherlands.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Additional information

The Netherlands organization for scientific research (NWO) is gratefully acknowledged for funding this project. This research was conducted while the author was supported by a PSYCHON-grant (560-267-011) from this organization. The author is obliged to Jos ten Berge and Pieter Kroonenberg.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Kiers, H.A.L. Hierarchical relations among three-way methods. Psychometrika 56, 449–470 (1991). https://doi.org/10.1007/BF02294485

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02294485

Key words

Navigation