I'm using lme4 package to run a mixed-effects logistic regression. I have a glmer model, and to assess its goodness of fit I created the binned residual plot. Unfortunately, 95% of the bins are not within the +/- 2 limit lines. So what to do now?
I tried to use the influence.ME package to detect the influential cases but I'm not sure whether I'm on the right track (and I get a Hessian convergence warning with the influence function).
Here is my model: model <- glmer(Unique ~ Time + Age_in_days + Condition + Time:Condition + (1+Time|Child_ID) + (1+Condition|Item), data=data, control=glmerControl(optimizer="bobyqa"), family = binomial(link=logit))
Time and condition are categorical. Age is continuous.
Any suggestions? Thank you for your time.