Message-ID: <49F1C530.9070108@biostat.ku.dk>
Date: 2009-04-24T13:57:04Z
From: Peter Dalgaard
Subject: Interpreting the results of Friedman test
In-Reply-To: <20090424134246.195060@gmx.net>
doerte.apelt at gmx.de wrote:
> Hello,
>
> Peter Dalgaard wrote:
>> Not necessarily, but the first suspicion one gets is that what you're
>> doing with Friedman is not equivalent to what you're doing with ANOVA,
>> could you show us the code and data (or an outline of it)?
>
> Please find attached the R script and the data input files.
> The values in dataForFriedmanTest.dat represent the geometric mean of
> the values of three repetitions for one observer and one condition.
>
> Am I doing something wrong there?
The .R file came as BIN file to me, so r-help may be missing it.
Anyways, Friedman's test is a replacement for a two-way ANOVA and you
are comparing it to a one-way analysis, and the latter is likely just wrong.
Try
anova(lm(AUC~as.factor(Condition)+as.factor(Observer),data=dataForANOVA))
or, maybe better
summary(aov(AUC ~ as.factor(Condition) + Error(as.factor(Observer) /
as.factor(Condition)), data=dataForANOVA))
> Thanks in advance
>
> Doerte
>
> PS: I'm not sure if it is possible to send the dat-files by mail.
> Therefore I renamed them to *.txt. Please rename back to *.dat.
--
O__ ---- Peter Dalgaard ?ster Farimagsgade 5, Entr.B
c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
(*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907