Hi all,

after browsing the archives for hours I'm still not sure about the proper 
analysis for my dataset.

I subjected each of about 50 critters (about 10 each in 5 distinct 
populations) to 4 consecutive treatments (exposure to increasing 
concentrations), with one measurement per treatment and individual.

I, of course, want to know, if there was a treatment and / or a population 
effect.

Since I'm dealing with 'paired' data, and also with a (expected) 'linear' 
response to a 'linear' treatment, how do I perform the proper analysis?

I tried something like:

summary(aov(y~group+treatment)
summary(lme(y~group+treatment)

but I don't see, how these formula would differentiate between a 'paired' 
or 'unpaired' experimental design. Somewhere in the archives it was mentioned 
that you could use the error-term to make that distinction (Baron-rpsych.pdf, 
pp.36-)

summary(aov(y~group+treatment+Error(?), na.action=na.omit)
summary(lme(y~group+treatment,random=~1|?,na.action=na.omit)

but, unfortunately, that level of sophistication is still beyond my 
statistical powers.


Any advice on the matter would be very much appreciated


Wolfgang

______________________________________________
R-help@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.

Reply via email to