Page 1 of 1

Error message "0 (non-NA) cases" linear regression

Posted: Tue Jul 04, 2023 12:24 pm
by flatliner
Hello,

I have an issue with running a linear regression analysis.

I have a dependent variable (ordinal scale, 1-2) and three independent variables (interval scale, 1-8) and coded and defined missing values as "99".

Whenever I am adding a second covariate to the model I am receiving the error message "0 (non-NA) cases".
When trying to run logistic regression it says "Variable 1 has no levels" (screenshot attached).

Did anyone experience the same issue and know how to solve it?

Many thanks in advance for your help!

Re: Error message "0 (non-NA) cases" linear regression

Posted: Wed Jul 05, 2023 12:54 am
by jonathon
this usually means that there are no or few combinations of 'Low info-sharing', 'No policy' and 'Short policy' where there aren't missing values.

jonathon

Re: Error message "0 (non-NA) cases" linear regression

Posted: Wed Jul 05, 2023 11:22 am
by flatliner
Hello Jonathan,

Thanks for the explanation.
I've identified the error and computed a new variable.