0

I hope I will get your precious help for my R problem.

I want to study the classroom, the teacher and the school effect/result on the pupil's success (or not) at school. I also want to know the effect of the age, the gender and if the pupils have or not repeated 1 year school on the pupil's success (or not).

The pupil's variables are (fixed effects levels) : the age (11 or 12), the gender (M or F), if they have repeated one year (yes or not), the success (yes or not). The other variables are (random effects levels) : the classroom they belong to (A and B), the teacher they have (K; L; M and N) and the school they belong to (C and D).

Firstly I am looking for the effect of the classroom, the effect of the teacher and the effect of the school on the pupil's success and secondly I want to know the effect of the age, the gender and if the pupils have (or not) repeated one year on the pupil's success.

How can I do my GLMM (hierarchical multilevel model) with R ?

Best,

  • 1
    It helps to give us a data frame for your problem, otherwise you'll have a lot of trouble applying the response. – RegressForward Jan 17 '14 at 19:16
  • (1) A lot of mixed models I see on this site seem impossibly complicated. I'd try simplify. (2) If you have a binary outcome (as implied), that dataset will have to be fairly large.(3) based on above you should be able to suggest possible R code – charles Jan 17 '14 at 21:03
  • Hi, I have my data.frame (.txt) but I can not attached file. How do you want me to give the data.frame ? – varin sacha Jan 17 '14 at 22:35
  • From what I read, your data has the same problem as in your previous question, too little number of random levels (4 is the maximum for teachers). As we have told you there, you cannot build a GLMM for that. Unless this changes, the question should be closed. – Henrik Jan 18 '14 at 08:28
  • Hi Henrik, Thanks for your answer, I thought that we had to count 4 random effect levels for "teacher", 2 random effect levels for "classroom" and 2 more random effect levels for "school", that means a total 8 random effect levels. It is not the case, as I can see ? – varin sacha Jan 18 '14 at 15:56

0 Answers0