|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
THREELEVEL ICCs with BAYES and catego... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
For a categorical (ordinal) variable, TYPE=THREELEVEL and Bayesian estimation is it correct to say that the within-level residual variance of a variable is 1? Much like the probit case. I'm interested in estimating the Level 3 ICC as: (Level 3 var)/(LEVEL 2 var + LEVEL 3 var +1) Is this correct? |
|
|
Yes. |
|
Back to top |
|
|