I've created a mixed effects logistic regression model using proportional data, and I'm having trouble interpreting my residual plot. I've been taught that when looking at the residuals, I should be worried about clumps and wedge-shapes. I'm not necessarily seeing either of those in my plot, but I am seeing a distinct downward trend in the residuals (screenshot below). I'm not sure what this means-can I still use this model, or does this trend indicate that some assumption is not met?

These are the residuals that I see if I write plot(Model1): enter image description here

This is what I get if I write plot(residuals(Model1): enter image description here

Does it matter which way I view the residuals? They look fine to me in the second photo, but the first photo has that linear trend.

I've uploaded a sample of my data to dropbox, in case that is helpful for troubleshooting this. The data can be found here

I'm doing a mixed-effects proportional logistic regression in r, using the glmer function in the lme4 package. I'm looking at the effect of temperature on behavior. The model I created is:

Model1<-glmer(AverageResponse~AverageTemperature+LightPeriod+(1|Location), data=SampleDataset, family=binomial)

Any help is greatly appreciated! If any additional information is necessary in order to answer this question, please let me know!


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.