[R] ANCOVA with defined error terms
hpdutra at yahoo.com
Thu Aug 20 04:46:42 CEST 2009
I am trying to run an ANCOVA with defined error terms. Thus I have to use
AOV and not lm.
my response variable is proportion of mice paw prints on track plates. These
plates were placed on plots that had vegetation and fruit manipulated to two
levels each (present or absent), and were sampled monthly for 14 months
(repeated measures). The fully crossed factor design was also blocked. My
sample size is just 3 and I cannot run the full model because I would ran
out of degrees of freedom.
I measured raccoon activity on these plots as a covariate. So I am running
the following model.
data = track))))
here is a sample of the data
plot veget fruit time block coon mice
p1 Vremoved Fintact 1 b1 8.605276544 26.67179738
p2 Vintact Fintact 1 b1 16.64929378 45
p3 Vremoved Fremoved 1 b1 22.12023855 26.67179738
p4 Vintact Fremoved 1 b1 16.64929378 41.57117579
p5 Vintact Fremoved 1 b2 30.73522506 38.09196076
p6 Vremoved Fintact 1 b2 22.12023855 26.67179738
p7 Vintact Fintact 1 b2 8.605276544 8.605276544
p8 Vremoved Fremoved 1 b2 8.605276544 8.605276544
p9 Vintact Fintact 1 b3 8.605276544 16.64929378
p10 Vintact Fremoved 1 b3 16.64929378 8.605276544
p11 Vremoved Fintact 1 b3 8.605276544 30.73522506
p12 Vremoved Fremoved 1 b3 8.605276544 16.64929378
p1 Vremoved Fintact 2 b1 22.12023855 41.57117579
p2 Vintact Fintact 2 b1 8.605276544 48.42882421
My question is. Is there a more simple way to run this mess using lm or not?
View this message in context: http://www.nabble.com/ANCOVA-with-defined-error-terms-tp25055311p25055311.html
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help