[R] ANOVA testing over nested MS term

[peter dalgaard]

On Mar 18, 2012, at 08:21 , shoreliner11 wrote:

> I'm still relatively new to R but was wondering if anyone could help me force
> R to compute the f-statistic etc using the  the nested term rather than the
> residual. In my particular case we were nesting a treatment effect by a
> replicated tank which was not non-significant enough (p>0.25)to be dropped
> from the statistical model.

Ouch!!!!! Don't _ever_ take high p-values as proof of absence of effect. 

> 
> Here's my code:
> summary(analysis1<-aov(Area~Treatment/Tank))
> 
> Area is the response variable, treatment is the main effect, nested within a
> replicated tank. Thanks so much for your help.
> 

I.e. Tanks are numbered within Treatments, so that Treatment 1, Tank 1 is distinct from Treatment 2, Tank 1?

Something like aov(Area~Treatment + Error(Treatment:Tank)) should do it. There's a wart in aov() causing this sort of model to throw a warning that "Error() model is singular", for reasons that I never quite fathomed. An explicit 

TrTa <- interaction(Treatment,Tank)

and then using Error(TrTa) gets rid of the warning and provides equivalent output.


> --
> View this message in context: http://r.789695.n4.nabble.com/ANOVA-testing-over-nested-MS-term-tp4481767p4481767.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> 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.

-- 
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com