I have a dataset with two variables, first variable is the test result, it attains values : 1,2,3,4,5. The other one is the golden standard with values 0,1 and missing. I am using the following model to find the effect of the test on the golden standard:
proc glimmix data=data;
class studyID PosResultDesc ;
model tbyn1 = PosResultDesc/ cl dist=binary or link=logit chisq;
random intercept / subject = studyID ;
run;
But the model does not converge. I attach the data, any thoughts what could be the reason?