[R] Interpreting the results of Friedman test

Peter Dalgaard P.Dalgaard at biostat.ku.dk
Fri Apr 24 15:57:04 CEST 2009


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             Øster 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




More information about the R-help mailing list