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.
summary(model<-(aov(mice~veget*fruit*time+(time*block)+(time*block*veget)+(time*block*fruit)+coon+Error(block/plot, 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. ______________________________________________ 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.
