This preview shows page 1. Sign up to view the full content.
Unformatted text preview: vourable responses very good and excellent are amalgamated.
Log-odds ratios involve comparisons between (‘poor’ versus ‘fair’), ( b there is perfect homogeneity within the categories
being collapsed), typically resulting in a considerable
loss of statistical power. Although several statistical
models for ordinal responses have been proposed, their
utilization in the epidemiological and biomedical literature has been minimal.
The purpose of this paper is twofold: first, to provide
a synthesized review of models for analysing data with
ordinal responses, and second, to evaluate their usefulness in epidemiological research, with particular emphasis on model formulation, interpretation of model
coefficients, and their implications. Ordinal models
that are considered include (1) cumulative logit or
the ‘grouped continuous’ model,3,4 (2) continuationmodel,5 (3) constrained and unconstrained partial proportional odds models,6 (4) polytomous logistic model,7,8
(5) adjacent-category logistic model,9 and (6) stereotype logistic model.10 The development of each model
is described in detail,...
View Full Document
This document was uploaded on 02/25/2014.
- Spring '11