I have several questions in my survey with a "not applicable" category. When I want to calculate with these questions I want to exclude these answers. I do not want to use "select cases" because I will lose respondents because different respondents answer "not applicable" on different questions.
I think these "not applicable" answers (with value 0) will cause problems when calculating Crombach alpha's with these questions. Therefore I have two questions:
- is it true that not applicable will cause problems for Crombach Alpha?
- if so, what might be a solution?
Thanks in advance, Esther