I have coded the responses with 1=yes 2=don't know 0=no. And 0= strongly agree ect. Having numerical data has allowed me to run a spearman's one tailed correlation test against 2 variables.
Is this the appropriate thing to do and will the results be valid?