[R-sig-ME] Simulating multilevel dataset

[Stuart Luppescu]

On Fri, 2013-07-05 at 16:53 +0000, Thompson,Paul wrote:
> You indicate that you have teacher level data and the school level data. You do not mention data about conditionality. How can you model it if you do not have some relationship to model? In addition, you do not specify the manner in which error is incorporated into the model.

The model would look something like this (in lmer notation): meas ~ var1.indic + var2.indic + (var1.indic|school) + (var2.indic|school)
Does this address your questions?

> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Stuart Luppescu
> Sent: Friday, July 05, 2013 11:04 AM
> To: r-sig-mixed-models
> Subject: [R-sig-ME] Simulating multilevel dataset
> 
> Hello, I'm trying to simulate a multilevel dataset representing teachers within schools. To generate the data I'm using parameters from actual data. I have the means, variances and covariances of teachers on two variables; and the means, variances and covariances of the school-level means for those two variables. I can generate the teacher-level data, and the school-level data using rmvnorm() with no problem, but what I can't figure out is how to generate teacher-level data conditional on the school-level generated data. 
> 
> Can anyone give me a clue?
> 
> TIA. 
> --
> Stuart Luppescu <slu at ccsr.uchicago.edu>
> University of Chicago
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> 
> -----------------------------------------------------------------------
> Confidentiality Notice: This e-mail message, including any attachments,
> is for the sole use of the intended recipient(s) and may contain
> privileged and confidential information.  Any unauthorized review, use,
> disclosure or distribution is prohibited.  If you are not the intended
> recipient, please contact the sender by reply e-mail and destroy
> all copies of the original message.
> 


-- 
Stuart Luppescu <slu at ccsr.uchicago.edu>
University of Chicago