I've searched the web deep and wide for an answer to my question. But there is little consistency to be found, which of course makes me very confused and thus leads me to asking here.

I have a binary dependent variable 0/1, however my independent variables are mainly likert scale and some are binary. I understand that keeping everything constant and the binary option is present it will increase the odds by x, but how do I interpret the likert scale?

My findings so far is that this is not possible, you should multiply it with a number in the scale or you should multiply it with itself.

E.g. I my scale is 0,1,2 where 0 = No, 1 = Don't know, 2 = Yes

Is it then appropriate to say that the odds when 0 is present is x, and to find yes I take the odds times * 2 or should it be odds times * odds times * odds times?

I'm doing this with statsmodels in python


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.