R help - Nov 2010 - repeated measure test

Hi,

This is a question regarding technique rather than an R specific issue.  I
have been asked to evaluate a 30+ year long term continuous survey of bird
presence/absence data that has an associated ocular estimate of the
vegetation community percent coverage.  The data are organized by
subpopulations (5), and by year ( 1991 - present).  We are interested in
gaining understanding on whether bird presence is 1) related to vegetation
community cover and 2) whether that relationship is variable over the
period of record.  Ultimately I can associate management to this, but that
data has not be pulled together yet for this survey.

My first thought was to simply look to a clustering and nmds to examine
which sites were more similar given the vegetation community.    Following
this I had intended to repeat that analysis using year as a grouping
variable to see if the clusters change.   This would then provide the basis
for a trajectory analysis.

Ok, so that might be rather very simple, I'm now considering a repeated
measures approach, but do not have much experience here.  Since every site
is the same in all years, what changes is the bird count and perhaps the
vegetation community coverage estimates.

Can someone let me know if this alternate approach is a better approach and
how to conduct the analysis in R?

Thanks very much.  If additional information is needed I can provide it,
but I didn't want to fill mail boxes with unnecessary info at this time.

Steve

Steve Friedman Ph. D.
Spatial Statistical Analyst
Everglades and Dry Tortugas National Park
950 N Krome Ave (3rd Floor)
Homestead, Florida 33034

Steve_Friedman at nps.gov
Office (305) 224 - 4282
Fax     (305) 224 - 4147

Steve:

Obviously, you would do better to consult a local statistician.
Presumably, you have none, so you would probably do better posting on
appropriate R SIGs -- e.g. R-sig-mixed-models, R-sig-ecology and the
liike. R-help is really not designed for this (although good souls
sometimes respond). In a similar vein, you might try seacrhing the net
for non-R statistical lists that might be helpful.

-- Bert

On Fri, Nov 12, 2010 at 5:40 AM,  <Steve_Friedman at nps.gov>
wrote:
>
> Hi,
>
> This is a question regarding technique rather than an R specific issue. ?I
> have been asked to evaluate a 30+ year long term continuous survey of bird
> presence/absence data that has an associated ocular estimate of the
> vegetation community percent coverage. ?The data are organized by
> subpopulations (5), and by year ( 1991 - present). ?We are interested in
> gaining understanding on whether bird presence is 1) related to vegetation
> community cover and 2) whether that relationship is variable over the
> period of record. ?Ultimately I can associate management to this, but that
> data has not be pulled together yet for this survey.
>
> My first thought was to simply look to a clustering and nmds to examine
> which sites were more similar given the vegetation community. ? ?Following
> this I had intended to repeat that analysis using year as a grouping
> variable to see if the clusters change. ? This would then provide the basis
> for a trajectory analysis.
>
> Ok, so that might be rather very simple, I'm now considering a repeated
> measures approach, but do not have much experience here. ?Since every site
> is the same in all years, what changes is the bird count and perhaps the
> vegetation community coverage estimates.
>
> Can someone let me know if this alternate approach is a better approach and
> how to conduct the analysis in R?
>
> Thanks very much. ?If additional information is needed I can provide it,
> but I didn't want to fill mail boxes with unnecessary info at this
time.
>
> Steve
>
> Steve Friedman Ph. D.
> Spatial Statistical Analyst
> Everglades and Dry Tortugas National Park
> 950 N Krome Ave (3rd Floor)
> Homestead, Florida 33034
>
> Steve_Friedman at nps.gov
> Office (305) 224 - 4282
> Fax ? ? (305) 224 - 4147
>
> ______________________________________________
> 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