Categorical Data Analysis Using The SAS® System, 2nd Edition

9.2. Ordinal Response: Proportional Odds Model

9.2.1. Methodology

Consider the arthritis pain data in Table 9.1. Male and female subjects received an active or placebo treatment for their arthritis pain, and the subsequent extent of improvement was recorded as marked, some, or none (Koch and Edwards 1988).

Table 9.1. Arthritis Data

		Improvement
Sex	Treatment	Marked	Some	None	Total
Female	Active	16	5	6	27
Female	Placebo	6	7	19	32
Male	Active	5	2	7	14
Male	Placebo	1	0	10	11

One possible analysis strategy is to create a dichotomous response variable by combining two of the response categories, basing a model on either Pr{marked improvement} versus Pr{some or no improvement} or Pr{marked or some improvement} versus Pr{no improvement}. However, since there is a natural ordering ...