lmer problem

<No dataset - too big>
Hi all,
I have a dataset (attached) which shows the score of pupils on different subjects. For example, pupil 13 may have been tested on subjects 1 (Greek), subject 4 (History) and 5 (Latin) whereas pupil 124 may have been tested on subjects 1 (Greek), 38 (Physics) and 19 (Chemistry). All pupils took Greek (subject with code 1). I attach the file with some data. Below, I show some of the R commands that I used. My intention is to estimate which of the subjects was more difficlut, and to estimate the % of variance because of pupils and because of subjects. Please give me some help, also send me any commands you may want to suggest and some explanations. I realized that SPSS 15 has included a new mixed models component, is it much better than R?


library(RODBC)
channel <- odbcConnectExcel("C:/JASON/PROJECTS/vathmoi2007/alldata.xls")
channel <- odbcConnectExcel("C:\smalldata.xls")
sqlTables(channel)
sqlTables(channel, errors = FALSE, as.is = TRUE)
Dataset = sqlQuery(channel, paste("select * from [data$]"))
Dataset$mathima <- factor(Dataset$mathima)
Dataset$app_aa <- factor(Dataset$app_aa)

note: app_aa is the code of the pupil

this is the model for the analysis I used, but how do I show that the pupils take more than one test?
mod2 <- lmer(arxiki ~ mathima + (1|app_aa), Dataset)

how about mod3 <- lmer(arxiki ~ 1 + (1|mathima) + (1|mathima:app_aa), Dataset)


Thanks

Jason
 
Dr. Iasonas Lamprianou
Department of Education
The University of Manchester
Oxford Road, Manchester M13 9PL, UK
Tel. 0044 161 275 3485
iasonas.lamprianou at manchester.ac.uk


----- Original Message ----
From: "r-sig-mixed-models-request at r-project.org" <r-sig-mixed-models-request at r-project.org>
To: r-sig-mixed-models at r-project.org
Sent: Friday, 20 July, 2007 1:00:13 PM
Subject: R-sig-mixed-models Digest, Vol 7, Issue 13


Send R-sig-mixed-models mailing list submissions to
    r-sig-mixed-models at r-project.org

To subscribe or unsubscribe via the World Wide Web, visit
    https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
or, via email, send a message with subject or body 'help' to
    r-sig-mixed-models-request at r-project.org

You can reach the person managing the list at
    r-sig-mixed-models-owner at r-project.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-sig-mixed-models digest..."


Today's Topics:

   1. multinomial mixed effects models (H. Skaug)


----------------------------------------------------------------------

Message: 1
Date: Thu, 19 Jul 2007 10:21:50 +0200
From: "H. Skaug" <hskaug at gmail.com>
Subject: [R-sig-ME] multinomial mixed effects models
To: r-sig-mixed-models at r-project.org
Message-ID:
    <ed96c8240707190121u5a4f2a92h6354e01dcca4b952 at mail.gmail.com>
Content-Type: text/plain

Hi,

If you are willing to move outside R,
mixed effects multinomial models can be fit with the
commercial software AD Model Builder:

http://otter-rsch.com/admbre/admbre.html

Ordered and unordered categorical responses can
be handled, and there is no limit on the number
of nesting levels (in principle). For an example see:

http://otter-rsch.com/admbre/examples/socatt/socatt.html

It is easy to call an AD Model Builder program from R:

http://otter-rsch.com/admbre/examples/glmmadmb/glmmADMB.html

There obviously are limitations on the size of the model,
and I will be happy to clarify if your problem at hand is within reach.

hans

Hello!

I and several of my colleagues are wondering whether it is possible to
use any of the methods of lme4 as it exists now to fit a mixed effects
model with a response variable drawn from a multinomial distribution.
glm does not include a multinomial family, so if it is possible to
accomplish this I'm not sure how to do so.  Packages that do allow
multinomial response variables (like multinomRob) don't seem to allow
for the inclusion of random effects.

If it is not currently possible to fit a data set with a categorical
dependent variable with more than two levels, might this be possible in
the forthcoming update to lme4?

Finally, if it isn't possible now and won't be in the next version of
the package either, would someone be willing to explain the conceptual
or technical difficulties associated with including a response variable
from a multinomial distribution in a mixed effects model?

Thanks for any help,
/au

------------------------------

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


End of R-sig-mixed-models Digest, Vol 7, Issue 13
*************************************************


      ___________________________________________________________