|
Procedures for analysing data measured at the nominal level only. The goal, as in regression, is to fit an equation which predicts the values in a contingency table (i.e. the entries in each cell) in terms of the values of independent variables (which may also be measured at the nominal level only). The terms in the model, presented in logarithmic form, are the deviations for the relevant cell (defined by the values of the independent variables) from a control value, usually the grand mean for the entire sample. (Cf. categorical data analysis.)Â (RJJ)
Suggested Reading O\'Brien, L. 1990: The statistical analysis of contingency table designs. Concepts and techniques in modern geography 51. Norwich: Environmental Publications. O\'Brien, L. 1992: Introducing quantitative geography: measurement, methods and generalised linear models. London and New York: Routledge. |
|