I have an independent variable that has 3 groups - 'Yes' 'No' and 'Unknown'. I have decided that I can assume the 'Unknowns' as 'No'. I want to run sensitivity analysis to test whether making this assumption is ok and the 'Yes' 'No' independent variables can be included in logistic regression.

What would be the most appropriate method for this?

  • 1
    $\begingroup$ What else is in the data other than this variable? Why wouldn't you just want to take the variable as having three categories? What is the subject matter justification for treating "Unknown" as "No"? All these play important roles when addressing your question. $\endgroup$ – Lewian May 22 '20 at 12:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.