Either base items on within-condition symptomology cluster (conduct factor analysis / IRT accordingly), or conduct it on all items or specific subsets that are theoretically / substantively meaningful.

Hope this helps

Matt

Thanks, much Mattew! One problem is that I will lose most if the information by basing the items on the symptom clusters because most subjects only have a few of these symptoms. For IRT it makes sense but for FA it does not. The case-wise deletion will remove most of the data. One initial thought was to merge the two questions in one as wxplained in my RG question. Do you have any comment on the merges question solution?

Hello Saeid,

It's not difficult to recode the answers from the two questions into a single variable (as your query suggests); I see no risk to response validity as a result. I agree that the scale is not interval (from 0 on); polytomous IRT would be a sensible option, presuming that you have a set of comparable symptoms to consider.

But, you can still compare cases with, without a symptom/diagnosis fruitfully either way.

Good luck with your work.

Thanks, David. I am happy to hear that you confirm the strategy of meeging the gated queation with the sunsequent questions. I appreciate if you provide also a referece that supports the idea. Regards, saeid