Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: crossed 4-level linear analysis


From   "Hommes Juliette (EDUC)" <juliette.hommes@maastrichtuniversity.nl>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: crossed 4-level linear analysis
Date   Tue, 12 Jul 2011 18:11:49 +0200

Hi Neil, 

Thank you so much for your help. Indeed I've been able to do so now using Glamm. 
The unfortunate difficulty is the crossed nature with so many groups (n=30 in module 1 to n=33 in module 2). This creates so many random intercepts that it's to difficult to analyse. So far I've thus analysed the data within the (seperate) modules (=no variation in group composition). Do you think there is another way of including module and groups within the same analysis?

Thanks so much for your help again.

Juliette

Op 12 jul 2011, om 11:43 heeft Nick Darson het volgende geschreven:

Juliette,

I haven't used xtmixed much myself, but according to these threats it
might be possible to model four levels:

http://statalist.1588530.n2.nabble.com/st-Multilevel-models-Multilevel-mixed-effects-model-td3897455.html




On Tue, Jul 12, 2011 at 7:34 PM, Neil Shephard <nshephard@gmail.com> wrote:
> On Tue, Jul 12, 2011 at 10:24 AM, Hommes Juliette (EDUC)
> <juliette.hommes@maastrichtuniversity.nl> wrote:
>> L.S.,
>> 
>> With great enthusiasm I've worked with STATA but now I've come across a problem that I cannot resolve so far. Therefore I hope you can give me a hand.
>> 
>> I've collected repeated measurements (2) in two modules. Participants enrolled in the first module and were divided in groups randomly (n=30). After finishing the first module, the participants are again randomly assigned to groups (n=32). As such I've got four levels in my data structure with "crossed"data among the students/group levels: 1) Repeated measures, 2) Modules, 3) Students, 4) Groups.
>> 
>> I've been using the xtmixed command, but I found that it could only be used for three levelled datasets (and/or crossed data). Do you have an idea to analyse the four levels in one model and account for the crossed nature of the data in STATA?
> 
> I'm not certain whether it can handle what you want, but you might
> want to see if -gllamm- is of use here (http://www.gllamm.org/).
> 
> Within Stata you can get a description using -ssc desc gllamm-
> 
> Neil
> 
> --
> “Truth in science can be defined as the working hypothesis best suited
> to open the way to the next better one.” - Konrad Lorenz
> 
> Email - nshephard@gmail.com
> Website - http://kimura.no-ip.org/
> Photos - http://www.flickr.com/photos/slackline/
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index