Hi all,
I have data from an experiment with 3 independent variables, 2 are
within and 1 is between. In addition to the dependent variable, I have
a covariate that is a single measure per subject. Below I provide an
example generated data set and my approach to implementing the ANCOVA.
However the output confuses me; why does the covariate only appear in
the first strata? Presumably it should appear in every strata in which
there can be between-Ss effects or interactions with between-Ss
effects, no?
#generate data
set.seed(1)
a=rbind(
expand.grid(
id=1:20
,iv1 = 1:2
,iv2 = 1:2
)
)
a$group = factor(a$id<11)
a$dv = rnorm(length(a[,1]))
a$covariate = NA
for(i in unique(a$id)){
a$covariate[a$id==i]= rnorm(1)
}
#make sure id, iv1, and iv2 are factorized
a$id=factor(a$id)
a$iv1=factor(a$iv1)
a$iv2=factor(a$iv2)
#run ANCOVA
covariate_aov = aov(dv~covariate+group*iv1*iv2+Error(id/(iv1*iv2)),data=a)
summary(covariate_aov)
--
Mike Lawrence
Graduate Student
Department of Psychology
Dalhousie University
www.thatmike.com
Looking to arrange a meeting? Check my public calendar:
http://www.thatmike.com/mikes-public-calendar
~ Certainty is folly... I think. ~
I'm surprised there were no takers on this query; I thought it would be an easy answer, particularly where I provided example data set and code. Did my request run afoul of the list etiquette? On Tue, Feb 10, 2009 at 5:12 PM, Mike Lawrence <mike at thatmike.com> wrote:> Hi all, > > I have data from an experiment with 3 independent variables, 2 are > within and 1 is between. In addition to the dependent variable, I have > a covariate that is a single measure per subject. Below I provide an > example generated data set and my approach to implementing the ANCOVA. > However the output confuses me; why does the covariate only appear in > the first strata? Presumably it should appear in every strata in which > there can be between-Ss effects or interactions with between-Ss > effects, no? > > #generate data > set.seed(1) > a=rbind( > expand.grid( > id=1:20 > ,iv1 = 1:2 > ,iv2 = 1:2 > ) > ) > a$group = factor(a$id<11) > a$dv = rnorm(length(a[,1])) > a$covariate = NA > for(i in unique(a$id)){ > a$covariate[a$id==i]= rnorm(1) > } > > #make sure id, iv1, and iv2 are factorized > a$id=factor(a$id) > a$iv1=factor(a$iv1) > a$iv2=factor(a$iv2) > > #run ANCOVA > covariate_aov = aov(dv~covariate+group*iv1*iv2+Error(id/(iv1*iv2)),data=a) > summary(covariate_aov) > > > -- > Mike Lawrence > Graduate Student > Department of Psychology > Dalhousie University > www.thatmike.com > > Looking to arrange a meeting? Check my public calendar: > http://www.thatmike.com/mikes-public-calendar > > ~ Certainty is folly... I think. ~ >-- Mike Lawrence Graduate Student Department of Psychology Dalhousie University www.thatmike.com Looking to arrange a meeting? Check my public calendar: http://www.thatmike.com/mikes-public-calendar ~ Certainty is folly... I think. ~
Hi Mike,>> I'm surprised there were no takers on this query; I thought it would >> be an easy answer, particularly where I provided example data set and >> code.The following simplification should help you to answer your own question. covariate_aov = aov(dv~(covariate+group+iv1+iv2)^2,data=a) You can find out more about how to specify models in Chap. 11 of the manual "An Introduction to R.">> Did my request run afoul of the list etiquette?I don't believe so. You can never be sure that someone who knew the answer saw your first posting. Regards, Mark. Mike Lawrence-7 wrote:> > Hi all, > > I have data from an experiment with 3 independent variables, 2 are > within and 1 is between. In addition to the dependent variable, I have > a covariate that is a single measure per subject. Below I provide an > example generated data set and my approach to implementing the ANCOVA. > However the output confuses me; why does the covariate only appear in > the first strata? Presumably it should appear in every strata in which > there can be between-Ss effects or interactions with between-Ss > effects, no? > > #generate data > set.seed(1) > a=rbind( > expand.grid( > id=1:20 > ,iv1 = 1:2 > ,iv2 = 1:2 > ) > ) > a$group = factor(a$id<11) > a$dv = rnorm(length(a[,1])) > a$covariate = NA > for(i in unique(a$id)){ > a$covariate[a$id==i]= rnorm(1) > } > > #make sure id, iv1, and iv2 are factorized > a$id=factor(a$id) > a$iv1=factor(a$iv1) > a$iv2=factor(a$iv2) > > #run ANCOVA > covariate_aov = aov(dv~covariate+group*iv1*iv2+Error(id/(iv1*iv2)),data=a) > summary(covariate_aov) > > > -- > Mike Lawrence > Graduate Student > Department of Psychology > Dalhousie University > www.thatmike.com > > Looking to arrange a meeting? Check my public calendar: > http://www.thatmike.com/mikes-public-calendar > > ~ Certainty is folly... I think. ~ > > ______________________________________________ > 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. > >-- View this message in context: http://www.nabble.com/Mixed-ANCOVA-with-between-Ss-covariate--tp21942605p21976002.html Sent from the R help mailing list archive at Nabble.com.