[R] repeated measures anova
gunter.berton at gene.com
Mon Feb 18 15:26:53 CET 2013
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
On Mon, Feb 18, 2013 at 2:38 AM, Robert D Kinley <rdkinley at gmail.com> wrote:
> 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
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Genentech Nonclinical Biostatistics
Internal Contact Info:
More information about the R-help