R help - Feb 2013 - repeated measures anova

Hi

I'm having difficulty working out how to get what I think is the
appropriate partitioning of variability in a repeated measures setup.

I have G=5 treatment-groups, each containing n=6 subjects, and a response
is measured on each subject on  t=4 occasions.

I think the anova degrees of freedom should partition as follows -

Between-subjects:   G*n-1=29

   [ between-groups: g-1 = 4 ,

     between-subjects residual G*(n-1) = 25 ]

Within-subjects:   G*n*(t-1)=90

   [ between-occasions t-1=3  ,

     occasion:group interaction (G-1)*(t-1) = 12 ,

     within-subject residual G*(n-1)*(t-1)=75 ]

What's the best way to tackle this analysis in R * ?

Hope you can help

      cheers              Bob Kinley
*[ R 2.15.2 , windows xp ]

	[[alternative HTML version deleted]]

Hi

I'm having difficulty working out how to get what I think is the appropriate
partitioning of variability in a repeated measures setup.

I have G=5 treatment-groups, each containing n=6 subjects, and a response is
measured on each subject on  t=4 occasions.

I think the anova degrees of freedom should partition as follows - 

Between-subjects:   G*n-1=29
   [ between-groups: g-1 = 4 ,  
     between-subjects residual G*(n-1) = 25 ]

Within-subjects:   G*n*(t-1)=90
   [ between-occasions t-1=3  ,  
     occasion:group interaction (G-1)*(t-1) = 12 , 
     within-subject residual G*(n-1)*(t-1)=75 ]

What's the best way to tackle this analysis in R * ?

Hope you can help

      cheers              Bob Kinley

*[ R 2.15.2 , windows xp ]

1. Please do not multiple post.

2. One approach would be mixed effects models. See the nlme and lme4 packages.

3. But there are a host of others (including, perhaps, just ?aov.).

4. This is a statistical question, and off topic here. Post on the
r-sig-mixed-effects models list instead or s statistical list like
stats.stackexchange.com.

-- Bert

On Mon, Feb 18, 2013 at 2:38 AM, Robert D Kinley <rdkinley at gmail.com>
wrote:
>  Hi
>
> I'm having difficulty working out how to get what I think is the
> appropriate partitioning of variability in a repeated measures setup.
>
> I have G=5 treatment-groups, each containing n=6 subjects, and a response
> is measured on each subject on  t=4 occasions.
>
> I think the anova degrees of freedom should partition as follows -
>
> Between-subjects:   G*n-1=29
>
>    [ between-groups: g-1 = 4 ,
>
>      between-subjects residual G*(n-1) = 25 ]
>
> Within-subjects:   G*n*(t-1)=90
>
>    [ between-occasions t-1=3  ,
>
>      occasion:group interaction (G-1)*(t-1) = 12 ,
>
>      within-subject residual G*(n-1)*(t-1)=75 ]
>
> What's the best way to tackle this analysis in R * ?
>
> Hope you can help
>
>       cheers              Bob Kinley
> *[ R 2.15.2 , windows xp ]
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.


-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm