Graphical representation of quantified categorical data: its inherent problems

https://doi.org/10.1016/0378-3758(94)00013-LGet rights and content

Abstract

Given a two-way incidence table, its rows and columns are assigned values so that the regression of rows on columns and the regression of columns on rows are simultaneously linear. In this quantification, it is the general practice to plot weights for rows and columns in the same graph for the purpose of interpreting the data structure. This joint graphical display of quantified categorical data, however, have a number of problems. The present study is concerned with two objectives: one to identify some of those problems that the researcher should be aware of, and the other to offer some alternative ideas.

References (15)

  • J.P. Benzécri

    L'Analyse des Données: II. L'Analyse des Correspondances

    (1973)
  • J.D. Carroll et al.

    Interpoint distance comparisons in correspondence analysis

    J. Marketing Res.

    (1986)
  • J.D. Carroll et al.

    Comparing interpoint distances in correspondence analysis: a clarification

    J. Marketing Res.

    (1987)
  • J.D. Carroll et al.

    Reply to Greenacre's commentary on the Carroll-Green-Schaffer scaling of two-way correspondence analysis solutions

    J. Marketing Res.

    (1989)
  • A. Gifi

    Nonlinear Multivariate Analysis

    (1990)
  • M.J. Greenacre

    The Carroll-Green-Schaffer scaling in correspondence analysis: a theoretical and empirical appraisal

    J. Marketing Res.

    (1989)
  • L. Guttman

    The quantification of a class of attributes: a theory and method of scale construction

There are more references available in the full text version of this article.

Cited by (0)

View full text
Copyright © 1995 Published by Elsevier B.V.