dear all, I want to perform a posthoc test for my ANCOVA: a1<-aov(seeds~treatment*length) With summary(glht(a1, linfct = mcp(treatment = "Tukey"))) R tells me: "covariate interactions found -- please choose appropriate contrast" How do I build these contrasts? Ideally, I would like to have the posthoc test for the ANCOVA including a block-effect a2<-aov(seeds~treatment*length+Error(site)) How do I make a posthoc test here? Thanks for any comments Walter -- ***** Dr. Walter Durka Department Bioz?noseforschung Department of community ecology Helmholtz-Zentrum f?r Umweltforschung GmbH - UFZ Helmholtz Centre for Environmental Research - UFZ Theodor-Lieser-Str. 4 / 06120 Halle / Germany walter.durka at ufz.de / http://www.ufz.de/index.php?en=798 phone +49 345 558 5314 / Fax +49 345 558 5329 +++++++++++++++++++++++++++++++++++++++++++++++++ Das UFZ hat einen neuen Namen: Helmholtz-Zentrum f?r Umweltforschung GmbH - UFZ The UFZ has a new name: Helmholtz Centre for Environmental Research - UFZ +++++++++++++++++++++++++++++++++++++++++++++++++
On Tue, 9 Jan 2007, Walter Durka wrote:> dear all, >Walter, _please_ cc questions on contributed packages to the maintainer, since not everybody follows r-help closely...> I want to perform a posthoc test for my ANCOVA: > a1<-aov(seeds~treatment*length) > > With > summary(glht(a1, linfct = mcp(treatment = "Tukey"))) > R tells me: "covariate interactions found -- please choose appropriate > contrast" >one needs to specify a certain value for `length' which, I assume, is a numeric covariate, right? The current interface doesn't support this (this on my to-do-list), however, you can set up the matrix of linear functions by yourself (contact me privately if you have problems to do that).> How do I build these contrasts? > > Ideally, I would like to have the posthoc test for the ANCOVA including > a block-effect > a2<-aov(seeds~treatment*length+Error(site)) > > How do I make a posthoc test here? >its on the to-do-list as well :-( Torsten> Thanks for any comments > Walter > > > -- > > ***** > Dr. Walter Durka > Department Bioz?noseforschung > Department of community ecology > > Helmholtz-Zentrum f?r Umweltforschung GmbH - UFZ > Helmholtz Centre for Environmental Research - UFZ > Theodor-Lieser-Str. 4 / 06120 Halle / Germany > > walter.durka at ufz.de / http://www.ufz.de/index.php?en=798 > phone +49 345 558 5314 / Fax +49 345 558 5329 > > +++++++++++++++++++++++++++++++++++++++++++++++++ > > Das UFZ hat einen neuen Namen: > Helmholtz-Zentrum f?r Umweltforschung GmbH - UFZ > > The UFZ has a new name: > Helmholtz Centre for Environmental Research - UFZ > > +++++++++++++++++++++++++++++++++++++++++++++++++ > > ______________________________________________ > R-help at stat.math.ethz.ch 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. > >
## The WoodEnergy example in package HH (available on CRAN) is similar
## to the example you asked about. I have two factors and a
## covariate, and the interaction between the factors and covariate is
## significant. I construct the multiple comparisons of factor Stove
## at each level of factor Wood at a specified value of the covariate.
## open a trellis.device() with history recording on.
## There are many plots and you will need to move among them.
require(HH)
source(paste(file.path(.path.package(package="HH")[1],
"scripts"),
"MMC.WoodEnergy-aov.R", sep="/"))
## This call prints an Error message that you can ignore because
## the relevant call is inside a try() expression.
anova(energy.aov.4) ## source() needs a print() statement, demo() doesn't.
source(paste(file.path(.path.package(package="HH")[1],
"scripts"),
"MMC.WoodEnergy.R", sep="/"))
## The contrast matrix is developed beginning on line 91 of file
## MMC.WoodEnergy.R
## The MMC (mean-mean multiple comparisons) plots are described in R
## with ?MMC. The MMC plots in R use the glht package.
## Both files are commented. Please read the comments.
## Those source() commands in HH_1.17 will be replaced by demo()
## commands in the next version of HH.
##
## demo("MMC.WoodEnergy-aov", package="HH")
## demo("MMC.WoodEnergy", package="HH")
## glht is not currently able to work with aov objects that have the
## Error() function. You will need to respecify your model formula
## using the terms() function. See the maiz.aov example in the ?MMC
## help page.