clustered data analysis

[kende jan]

Dear all,
 
I have an design of clustered data with ordinal dependent variable (Low, moderate, high) and 3 cluster variables and 4 subject  variables.
Which is the procedure to take account of clustered data to estimate my model ?
 
thank you in advance
 
Jan


       

       
               
___________________________________________________________________________
Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions !
Profitez des connaissances, des opinions et des expériences des internautes sur Yahoo! Questions/Réponses
http://fr.answers.yahoo.com

[Dan Zetu]

Jan:

It is not very clear to me what is the difference between cluster variables
and subject variables as you mention below. Nonetheless, if your variables
are ordinal it is typically accepted to treat them as continuous (although
some of this list may disagree) and use any common clustering methods
(hierarchical and/or k-means).

Dan

>From: kende jan <[hidden email]>
>Reply-To: kende jan <[hidden email]>
>To: [hidden email]
>Subject: clustered data analysis
>Date: Mon, 29 Jan 2007 15:49:45 +0000
>
>Dear all,
>
>I have an design of clustered data with ordinal dependent variable (Low,
>moderate, high) and 3 cluster variables and 4 subject  variables.
>Which is the procedure to take account of clustered data to estimate my
>model ?
>
>thank you in advance
>
>Jan
>
>
>
>
>
>
>___________________________________________________________________________
>Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions
>!
>Profitez des connaissances, des opinions et des expériences des internautes
>sur Yahoo! Questions/Réponses
>http://fr.answers.yahoo.com

_________________________________________________________________
Get in the mood for Valentine's Day. View photos, recipes and more on your
Live.com page.
http://www.live.com/?addTemplate=ValentinesDay&ocid=T001MSN30A0701

[Hector Maletta]

        You may use multi level modelling to take account of variables at
cluster level. MIXED MODEL is the SPSS command.

        Hector

        -----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de kende
jan
Enviado el: 29 January 2007 12:50
Para: [hidden email]
Asunto: clustered data analysis

        Dear all,

        I have an design of clustered data with ordinal dependent variable
(Low, moderate, high) and 3 cluster variables and 4 subject  variables.
        Which is the procedure to take account of clustered data to estimate
my model ?

        thank you in advance

        Jan







___________________________________________________________________________
        Découvrez une nouvelle façon d'obtenir des réponses à toutes vos
questions !
        Profitez des connaissances, des opinions et des expériences des
internautes sur Yahoo! Questions/Réponses
        http://fr.answers.yahoo.com

[statisticsdoc]

Stephen Brand
www.statisticsdoc.com



Jan,

If by clustered data, you mean nested data (e.g. students within classrooms
within schools), then you would need to carry out Hierarchical Linear
Modeling.

The ordinal nature of the dependent variable raises an issue about the
statistical package that you would use.  You need a package that will create
a "link function" (Raudenbush & Bryk, 2002) between the predictors and
different steps in the ordinal measure.  The HLM 6 package certainly allows
you to analyze an ordinal outcome.  While the SPSS MIXED procedure will
allow you to carry out a variety of HLM analyses, and is a very powerful
tool, it is not clear to me that it enables you to carry out an HLM analysis
with an ordinal outcome.

If your Low, Medium, High is simply a classification of a continuous
variable, you might want to consider an HLM analysis of the continuous form
of the variable, using the SPSS MIXED procedure.

HTH,

Stephen Brand

For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
kende jan
Sent: Monday, January 29, 2007 10:50 AM
To: [hidden email]
Subject: clustered data analysis


Dear all,

I have an design of clustered data with ordinal dependent variable (Low,
moderate, high) and 3 cluster variables and 4 subject  variables.
Which is the procedure to take account of clustered data to estimate my
model ?

thank you in advance

Jan






___________________________________________________________________________
Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions !
Profitez des connaissances, des opinions et des expériences des internautes
sur Yahoo! Questions/Réponses
http://fr.answers.yahoo.com