Applied Multivariate Data Analysis

Part of the series Springer Texts in Statistics pp 1-129

Contingency Tables

  • J. D. JobsonAffiliated withFaculty of Business, University of Alberta

* Final gross prices may vary according to local VAT.

Get Access


This chapter begins with an introduction for Volume II and then presents a survey of the techniques available for analyzing contingency tables. The introduction consists of a discussion of data matrices measurement scales and an outline of techniques presented in Volume II. The discussion of contingency tables begins in the second section with a review of bivariate analysis for two categorical random variables and includes a discussion of inference techniques for two-dimensional tables. The discussion of two-dimensional tables also includes an introduction to the use of loglinear models. The third section presents a discussion of the application of loglinear models to multidimensional tables based on the maximum likelihood approach to estimation. The logit model is also introduced as a special case of the loglinear model. The last section of the chapter outlines the weighted least squares approach to modeling categorical data. The weighted least squares approach affords a greater variety of models than the maximum likelihood method.