Question to lme4: Higher Level Response

Elisabeth,


You might find this brief stats.stackexchange.com answer (and comments) that I wrote useful:

http://stats.stackexchange.com/questions/169512/multilevel-model-with-responses-only-at-level-2/169585#169585


Happy to hear other perspectives on this issue.


Jake

It is better to send such questions to the R-SIG-Mixed-Models mailing list,
which I have taken the liberty of cc:ing on this response.  Many of the
readers of that list will be able to respond more quickly and knowledgeably
than I can.

On Tue, Dec 1, 2015 at 3:42 PM Elisabeth Schubach <
elisabeth.schubach at uni-jena.de> wrote:

>
> Dear Dr. Bates,
>
> I have got a question to lme4: Is it possible to fit a model with a
> response on a higher level by predictors of lower levels? Or do I then
> run into the problem of atomistic fallacy? And the only way is to turn
> the model around and fit the lower level response by higher level
> predictors? I understand that in that case, pinning avoids the
> ecological fallacy.
>
> Thank you and kind regards,
> Elisabeth Schubach
>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

Elisabeth,

That sort of model is more easily accomplished with structural equation modeling software. Check out the references below. 

Croon, M. A. (2007). Predicting group-level outcome variables from variables measured at the individual level: A latent variable multilevel model. Psychological Methods, 12(1), 45-57. doi:10.1037/1082-989X.12.1.45

L?dtke, O., Marsh, H. W., Robitzsch, A., Trautwein, U., Asparouhov, T., & Muth?n, B. (2008). The multilevel latent covariate model: A new, more reliable approach to group-level effects in contextual studies. Psychological Methods, 13(3), 203-229. doi:10.1037/a0012869

Marsh, H. W., L?dtke, O., Nagengast, B., Trautwein, U., Morin, A. J. S., Abduljabbar, A. S., & K?ller, O. (2012). Classroom climate and contextual effects: Conceptual and methodological issues in the evaluation of group-level effects. Educational Psychologist, 47(2), 106-124. doi:10.1080/00461520.2012.670488


Steven J. Pierce, Ph.D.
Associate Director
Center for Statistical Training & Consulting (CSTAT)
Michigan State University

-----Original Message-----
From: Douglas Bates [mailto:bates at stat.wisc.edu] 
Sent: Wednesday, December 02, 2015 12:49 PM
To: Elisabeth Schubach <elisabeth.schubach at uni-jena.de>
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Question to lme4: Higher Level Response

It is better to send such questions to the R-SIG-Mixed-Models mailing list,
which I have taken the liberty of cc:ing on this response.  Many of the
readers of that list will be able to respond more quickly and knowledgeably
than I can.

On Tue, Dec 1, 2015 at 3:42 PM Elisabeth Schubach <

Dear Dr. Bates,

I have got a question to lme4: Is it possible to fit a model with a
response on a higher level by predictors of lower levels? Or do I then
run into the problem of atomistic fallacy? And the only way is to turn
the model around and fit the lower level response by higher level
predictors? I understand that in that case, pinning avoids the
ecological fallacy.

Thank you and kind regards,
Elisabeth Schubach