We consider the statistical problem of analyzing the association between two categorical variables from cross-classified data. The focus is put on measures which enable one to study the possible dependencies at a local level and to assess whether the data support some more or less strong association model. Statistical inference is envisaged using an imprecise Dirichlet model; this model answers several difficulties of frequentist and Bayesian methods.
Keywords. Contingency Tables, Association, Logical model, Directional association model, Statistical inference, Upper and lower probabilities, IDM, Prior ignorance, Bayesian inference.
Authors addresses:
Laboratoire de Psychologie Environnementale
Universite Paris 5
71 avenue Edouard Vaillant
92774 - Boulogne-Billancourt Cedex
E-mail addresses:
| Jean-Marc Bernard | jmbernard@psycho.univ-paris5.fr |