LOGIT-MODELS AND RELATED QUASI-SYMMETRICAL LOG-LINEAR MODELS FOR COMPARING RESPONSES TO SIMILAR ITEMS IN A SURVEY

Suppose that subjects respond to a battery of questions (items) of a s imilar nature in a survey, with each item having the same categorical scale. This article discusses models that express logits for the respo nse distributions in terms of subject and item effects. The models, wh ich generalize the Rasch model, have interpretations referring to subj ect-specific comparisons of the items. Recent literature shows that on e can estimate item parameters using estimates of main effect paramete rs in corresponding quasi-symmetric log-linear models. We discuss this connection giving primary attention to ordinal response models using adjacent-category logits and cumulative logits. For the case of two it ems, we give expressions for models and corresponding parameter estima tes that are the basis of simple tests of marginal homogeneity for squ are ordinal contingency tables.