Rank order data

Rankings and partial rankings, with or without ties, can be conveniently analyzed using ISGROUPS=0.

Each row is an element to be ranked.

Each column is a set of rankings. In the item label, place any interesting demographics about the person doing the ranking.

Note: if every ranking set includes every element, and ties are not allowed, then elements can be columns and ranking sets as rows. ISGROUPS=0 is not required.

In general, we allow each ranking (column) to define its own "ranking scale". This is equivalent to the Partial Credit model.

Measures for the elements are obtained. Measures for the ranking sets are meaningless and can be ignored.

Fit statistics, DIF and DPF analysis, and contrast analysis of residuals are all highly informative.