6.22 True, or false?

a. One reason it is usually wise to treat an ordinal variable with methods that use the ordering is that in tests about effects, chi-squared statistics have smaller df values, so it is easier for them to be farther out in the tail and give small P-values; that is, the ordinal tests tend to be more powerful.

b. The cumulative logit model assumes that the response variable Y is ordinal;

it should not be used with nominal variables. By contrast, the baselinecategory logit model treats Y as nominal. It can be used with ordinal Y, but it then ignores the ordering information.

c. If political ideology tends to be mainly in the moderate category in New Zealand and mainly in the liberal and conservative categories in Australia, then the cumulative logit model with proportional odds assumption should fit well for comparing these countries.

d. Logistic regression for binary Y is a special case of the baseline-category logit and cumulative logit model with J = 2.