[Rasch] scaling ranked responses on tests

Donald Bacon dbacon at du.edu
Fri Sep 18 23:22:18 EST 2009

Hi all --

   I have some test data wherein respondents read a scenario and then rank the appropriateness of five different courses of action (ranked 1-5).  Our teaching team has agreed on the most appropriate ranking (the "answer key").  I'm thinking I should recode the ranks as pairwise comparisons, and then analyze the comparisons as dichotomous variables.  However, I don't think these will achieve local independence.  Is there a better way to do this analysis, or any published work on this topic that I can read and cite?

Thanks --

Don Bacon
Professor of Marketing
Daniels College of Business
University of Denver

Please consider the environment before you print

More information about the Rasch mailing list