I want to measure the association between gender and the number of assignments of each score in a questionnaire. However, I have a 2x6 reference table with numbers less than 5 in more than one box (over 20% of the boxes).
Looking for some solution I found the test of "Goodman and Kruskal's range", but I'm not sure if this is the best choice. My question is in the assumption that this test is used to measure the association between ordinal variables, and I have a nominal variable in case the gender is nominal.I would be happy if anyone could give me a little help on this.