I'm working on Rstudio and I have a question about residues in a binomial glm's case.

  • I used a glm to correct my response variable (germination (ger_b), encode 0, 1) by the light, because this variable presents variation inside my treatments.

  • Because I want to use treatment and light as explaining variable, I I use the residuals of this as new response variable instead of binary germination.

Here's my code:

C_ger_b <- glm(ger_b ~ lum, family=binomial(link=logit), data=d)
R_ger_b <- residuals(C_ger_b)

My question is : Can I use this new variable in a binomial glm or glmer, even if it's not binary anymore?

Here is my new response variable :

    Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
-1.59181 -1.25091  0.84093  0.06977  1.00200  1.25993

Here is my initial response variable :

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
   0.00    0.00    1.00    0.61    1.00    1.00

I hope my question isn't inappropriate, thank you for your time.


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.