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.
Unable to display preview. Download preview PDF.
Cited Literature and References
- 2.Andersen, Erling B. (1980).Discrete Statistical Models With Social Science Applications.Amsterdam: North-Holland Publishing Company.Google Scholar
- 4.Bishop, Yvonne M.M., Fienberg, Stephen E., and Holland, Paul W. (1975).Discrete Multivariate Analysis: Theory and Practice.Cambridge, Ma.: MIT Press.Google Scholar
- 5.Christensen, Ronald (1991).Log-Linear Models.New York: Springer-Verlag.Google Scholar
- 6.Everitt, B.S. (1977).The Analysis of Contingency Tables.London: Chapman and Hall.Google Scholar
- 12.Reynolds, H.T. (1977).The Analysis of Cross-Classifications.New York: The Free Press.Google Scholar
- 14.Upton, Graham J.G. (1978).The Analysis of Cross-Tabulated Data.New York: John Wiley and Sons.Google Scholar