Page 1 of 1
Problem with level 1 . variance
Posted: Mon Nov 01, 2021 2:50 am
by georgecorbett
Hi, I am running a three-level iterative measurement model with binary results (level 1, person level 2, area level 3). In my model I can't select the first level box when I add the cons because it doesn't appear, only the fixed parameters, level 2 and level 3 boxes appear.
How can I choose the level 1 variance. Thanks!
Re: Problem with level 1 . variance
Posted: Mon Nov 01, 2021 12:01 pm
by ChrisCharlton
If you are running a binary logit model then the variance is fixed at (π^2)/3 (≈3.29) rather than being an estimated parameter, and this is why you cannot add this to your model.
See the following FAQ for more information about variances:
Partitioning variation across levels
Re: Problem with level 1 . variance
Posted: Sun Jul 17, 2022 9:57 am
by Michael42
ChrisCharlton wrote: ↑Mon Nov 01, 2021 12:01 pm
If you are running a binary logit model then the variance is fixed at (π^2)/3 (≈3.29) rather than being an estimated parameter, and this is why you cannot add this to your model.
See the following FAQ for more information about variances:
Partitioning variation across levels
Hey, thanks! I am running a binary logit model and it really worked.