# [R] ANOVA

Spencer Graves spencer.graves at pdf.com
Tue Sep 9 12:23:51 CEST 2003

```To expand on Winfried's comments, you may wish to try the following
before considering multivariate:

anova(lm(OQtil4 ~ Tone, data=gesamt))

and similarly for your other response variables.  There are ways of
putting this kind of thing in a loop, storing only the p values or
whatever you want.  However, unless you need a routine to do this many
times, it's probably best to just do it like this and be done with it.

hope this helps.  spencer graves

Winfried Theis wrote:
> Hi Britta!
>
>
> On Tue, 2003-09-09 at 10:57, Britta Lintfert wrote:
> [snip]
>
>>
>>now I want to know, if there are significant differents in the variables
>>OQ, GO, SK, RC depending on Tone. This I can do with an ANOVA , or??
>
> Yes, this is correct.
>
>>But when I start aov in R I get the following message:
>>
>>anova1 <- aov(Tone ~  OQtil4*OQ0*GOtil4*GO0*SKrhsJ*SK0*RCrhsJ*RC0,
>>data=gesamt)
>>Warning message:
>>"-" not meaningful for factors in: Ops.factor(y, z\$residuals)
>>
>>What's wrong with my Data?
>
> Nothing. It is your call to aov that is wrong. Formulas in R are
> formulated as "Dependent variable" ~ "influences". So what you are
> trying to do is to explain Tone by all your other variables and their
> interactions. If your variables are independent you may simply go ahead
> and use aov on each of them. Look first for a book on multivariate
> linear models...
>
>
>>
>>Thanks for helping me
>
> You're welcome.
>
> Winfried
>
>>Britta
>>
>>
>>
>>--
>>Britta.Lintfert at IMS.Uni-Stuttgart.de
>>Institut für Maschinelle Sprachverarbeitung
>>http://www.ims.uni-stuttgart.de
>>der Universität Stuttgart               Tel.: 0049/711/121-1372
>>Azenbergstr. 12, D-70174 Stuttgart      Fax.: 0049/711/121-1366
>>
>>______________________________________________
>>R-help at stat.math.ethz.ch mailing list
>>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>

```