1
$\begingroup$

I have the following GLMM:

success ~ age + gender + group/task + (1 + group/task|school/subject), family = binomial

I want to know whether participants' probability to succeed in certain problem-solving tasks can be predicted by the type of task. I have 6 tasks which can be categorized into 3 groups (A, B, C) with 2 tasks in each group. Each participant received 3 tasks (one from A, one from B, one from C; combination and order counterbalanced). Cochran's Q- and post-hoc McNemar tests revealed that the three groups differ in their success rates: A is easier than B and C, and B and C are equally difficult. I used crosstabs to analyze whether the tasks within each group differ in their success rates and found that they are different for A and B.

Now I would like to do a comparison of all individual tasks (not just those within one category).

My question is: Is the equation above correct in terms of the fixed effects or is there any reason to include group as an extra fixed effect (e.g. to see whether task has an effect on top of the group effects found in the McNemar tests)? Would that be unnecessary?

What does it mean to include both group/task and group?

$\endgroup$
6
  • $\begingroup$ what does the group\task mean in the equation? The equation as it's written is confusing. You don't mention the school\subject. Which package you intend to use? $\endgroup$
    – Steve
    Sep 4, 2014 at 18:01
  • $\begingroup$ Hi Steve, group\task is meant to be "task nested within group". This is as far as I know the way it is written in R. The school/subject means that my study subjects are nested within schools. What else is confusing? Sorry about that. I intend to use lme4 in R. $\endgroup$
    – Eva
    Sep 5, 2014 at 21:33
  • $\begingroup$ So the \ is actually a /. Right? $\endgroup$
    – Steve
    Sep 8, 2014 at 10:12
  • $\begingroup$ Yes, of course. Sorry. I just edited the post. $\endgroup$
    – Eva
    Sep 8, 2014 at 17:14
  • $\begingroup$ I really don't know what should I answer first, its seems that you haven't read anything regarding the lme4 package, you should be able to answer your questions by reading even the manual of the package. In addition, I would suggest you have a look at lme4.r-forge.r-project.org . Your equation is wrong, but for more reasons than the one you ask. If you edit it a lot I could provide an answer, but it needs at least some basis of argument. $\endgroup$
    – Steve
    Sep 9, 2014 at 19:28

1 Answer 1

0
$\begingroup$

First, let's try to simulate some data with a similar structure of your problem's data. My understanding from your description is that we have something like this:

Say we have 4 schools with 8 subjects in each school, thus a total of 32 subjects. Each subject is posed with 3 tasks (one from each group), thus a total of 96 observations. As we have 6 different tasks, we will ask 16 times each task (6*16=96), so 32 tasks from each group (32*3=96).

n.schools <- 2
n.subject <- 20
n         <- n.subject*3
subject   <- rep(1:n.subject, 3)
school    <- rep(1:n.schools, (n/n.schools))

age    <- rnorm(n.subject, 30, 5)
age    <- rep(age,3)
gender <- sample(c(0,1),n.subject, replace=TRUE)
gender <- rep(gender,3)

A     <- c(rep(1,n/3),rep(0,2*n/3))
B     <- c(rep(0,n/3),rep(1,n/3), rep(0, n/3))
C     <- c(rep(0,2*n/3),rep(1,n/3))
group <- c(rep("A",n/3), rep("B",n/3),rep("C",n/3))
task  <- c(rep("a",n/6),rep("b",n/6),rep("c",n/6),rep("d",n/6),rep("e",n/6),rep("f",n/6))

u.subject <- rnorm((n/3), 0, 1)
u.school  <- rnorm(n.schools, 0 ,1)

lattent <- -20 + 0.5*age + 2*gender + 2*A -3*B -3.2*C + u.subject + u.school + rnorm(n,0,1)
pr      <- 1/(1+exp(-lattent))
success <- rbinom(n, 1, pr)

Now let's try a GLMM model as the one you propose:

library(lme4)
> glmer(success ~ age + gender + group/task + (1 + group/task | school/subject), family=binomial)
Error: number of observations (=60) < number of random effects (=360) 
for term (1 + group/task | subject:school); 
the random-effects parameters are probably unidentifiable

So what does this mean? In general, you need your mixed effects model to be identifiable, and one of the conditions for this to happen is $\sum_{i=1}^N (n_i-k) >0$, where $k$ is the number of random effects. In the case of balanced data, this can be written equivalently $N n_1 > N k$ or $n_1 > k$. In other words you need to have less random parameters than the number of observations in each cluster/group, subject in your case. This cannot be the case if you add the group/task term in the random part.

What group/task or school/subject means in the formula?

This is simply an equivalent way of writing group + group:task so you actually have group and its interaction with task. So the last question doesn't actually makes sense. The same principle is used when you put this term as a grouping factor in the right part of (1 | ...) which actually translates to (1| school) + (1 | school:subject) hence the one factor nested within the other.

So, I suggest you read Chapter 2 from Bates book and pay extra attention to model specification. It is really important to define what actually makes sense to include as fixed or random effects and what the grouping factors will be. This depends on how you designed the study and what you want to extract from your models.

$\endgroup$
9
  • $\begingroup$ Thanks for your efforts, Steve.I thought by modelling group/task I would tell the program that I have a nested structure, as it is the case with school and subject. I will read Bates' chapter and might come back to here if I have further questions. Thanks a lot for your help, it is greatly appreciated. $\endgroup$
    – Eva
    Sep 15, 2014 at 8:16
  • $\begingroup$ You can use them too as grouping factor, just be carefull about the levels and the structure of the random effects. If you think this answer helps and answers your question you can vote up and accept it. $\endgroup$
    – Steve
    Sep 15, 2014 at 9:30
  • $\begingroup$ Thanks, this is very helpful. Unfortunately I can't vote answers up yet (not enough reputation). $\endgroup$
    – Eva
    Sep 22, 2014 at 9:52
  • $\begingroup$ Thanks a gain for your help. I was wondering how I calculate the number of random effects in my model. So in your example, how do you get to the 360? $\endgroup$
    – Eva
    Oct 9, 2014 at 18:11
  • $\begingroup$ I read Bates' chapter and I see what he is doing with the random intercepts for nested variables, but the chapter does not cover random slopes for nested factors.? Is it theoretically - leaving aside for a moment the question of whether the model is identifiable - possible to include a random slope for a nested factor? For theoretical and empirical reasons I want to include a random slope for task on subject. And since task is nested within group, I thought I had to include it in the random effects structure as well. So, do I understand correctly that the "only" wrong thing in the ... $\endgroup$
    – Eva
    Oct 9, 2014 at 18:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.