Correspondence Analysis of Two-way Contingency Tables with Ordered Column Categories

  • Published : 1999.09.01

Abstract

Correspondence analysis is an exploratory method for two-way contingency tables intended to display the association pattern between row and column categories. It has been developed for several decades mainly in France and Japan and, nowadays, is popular worldwide. For the special case, however, that the column consists of ordered categories, correspondence analysis may yield ackward result so that it cannot be fully accommodated. That's because the row and column categories are assumed to be nominal in the ordinary correspondence analysis. The aim of this study is to develop the correspondence analysis for two-way contingency tables with ordered column categories. It is based on two specific algorithms,

Keywords

References

  1. Theory and Applications of Correspondence Analysis Greenacre, M.J.
  2. Correspondence Analysis in Practice Greenacre, M.J.
  3. Journal of the American Statistical Association v.82 The geometric interpretation of correspondence analysis Greenacre, M.J.;Hastie, T.
  4. Multivariate Descriptive Statistical Analysis Lebart, L.;Morineau, A.;Warwick, K.M.
  5. Multivarialte Analysis Mardia, K.V.;Kent, J.T.;Bibby, J.M.
  6. Analysis of Categorical Data: Dual Scaling and Its Applications Nishisato, S.
  7. Psychometrika v.40 Nonlinear programming approach to optimal scaling of partially ordered categories Nishisato, S.;Arri, P.S.
  8. The Korean Communications in Statistics v.4 Partial quantification in principal component analysis Suh, H.S.;Huh, M.H.
  9. Correspondence Analysis Specific to Certain Types of Categorical Data. Ph.D. Dissertaion. Department of Statistics Yang, K.S.