I'm looking for advice on how to best code two survey responses when I calculate inter-annotator agreement (krippendorff's alpha)
I have the code side of it down, it's the theory of how to deal with these responses that I'm not sure about.
The question is: do you find this text offensive? And the possible answers are 1, 2, 3, 4, 5, which all equate to yes, d which means don't know, and n which means no.
The way I see it, I have the following options:
'd' = 0 and n = '' (e.g. excluded)
'd' = 0 and 'n' = 0
'd' = '' and n = ''
PS: I'm working in python