Tucker2 as a Second-order Principal Component Analysis

  • Takashi Murakami
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)


Statistical properties of the Tucker2 (T2) model, a simplified version of three-mode principal component analysis (PCA), are investigated aiming at applications to the study of factor invariance. The T2 model is derived as a restricted form of second-order PCA in the situation comparing component loadings and component scores across occasions. Several statistical interpretations of coefficients obtained from the least squares algorithm of T2 are proposed, and several aspects of T2 are shown to be natural extensions of characteristics of classical PCA. A scale free formulation of T2 and a new derivation of the algorithm for the large sample case are also shown. The relationship with a generalized canonical correlation model is suggested.


Singular Value Decomposition Component Score Orthogonal Rotation Pattern Matrix Asymmetric Role 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Bloxom. B. (1984): Tucker’s three-mode factor analysis model. In: Research Methods for Multimode Data Analysis,Law, H.G.(eds.), 104–120, Praeger Publishers, New York.Google Scholar
  2. Cliff, X. (1966): Orthogonal rotation to congruence, Psychometrika, 31, 33–42.Google Scholar
  3. Gifi, A. (1990): Nonlinear Multivariate Analysis. Chichester: Wiley.Google Scholar
  4. Hiers, H.A.L. et al. (1992): An efficient algorithm for TUCKALS3 on data with large number of observation unit. Psychometrika. 57, 415–422.CrossRefGoogle Scholar
  5. Kroonenberg, P.M. (1983): Three-mode Principal Component Analysis, DSWO Press. Leiden.Google Scholar
  6. Kroonenberg. P.M. and De Leeuw. J. (1980): Principal component analysis of three-mode data by means of alternating least squares algorithm, Psychometrika, 45. 69–97.Google Scholar
  7. Meredith. W. & Millsap, R.E. (1985): On component analysis. Psychometrika, 50, 495–507. Meredith, W. and Tisak, J. (1982): Canonical analysis of longitudinal and repeated measures data with stationary weights. Psychometrika. 47, 47–67.Google Scholar
  8. Muralami, T. (1983): Quasi three-mode principal component analysis: A method for assessing the factor change, Behavìormetrika, 14. 27–48.CrossRefGoogle Scholar
  9. Nishisato, S. (1994): Elements of Dual Scaling: An Introduction to Practical Data Analysis, Lawrence Erlbaum, Hilsdale.Google Scholar
  10. Ten Berge, J.M.F. (1986): Some relationship between descriptive comparisons of components from different studies, Multivariate Behavioral Research, 21, 29–40.CrossRefGoogle Scholar
  11. Ten Berge. J._11.F. (1988): Generalized approach to the MAXBET problem and the MAXDIFF problem, with applications to canonical correlations. Psychometrika, 53, 487–494.MATHGoogle Scholar
  12. Ten Berge, J.M.F. (1993): Least Squares Optimization in Multivariate Analysis. DSWO Press, Leiden.Google Scholar
  13. Ten Berge. J.M.F. and Kiers, H.A.L. (1996): Optimality criteria for principal component analysis and generalizations. British Journal of Mathematical and Statistical Psychology. 49, 335–345.MathSciNetMATHCrossRefGoogle Scholar
  14. Tucker,. L.R. (1966): Some mathematical notes on three-mode factor analysis. Psychometrika, 31. 279–311.Google Scholar

Copyright information

© Springer Japan 1998

Authors and Affiliations

  • Takashi Murakami
    • 1
  1. 1.School of EducationNagoya University Furo-choChikusa-ku, NagoyaJapan

Personalised recommendations