[R] lme nesting/interaction advice

Andrew Robinson A.Robinson at ms.unimelb.edu.au
Mon May 12 12:16:26 CEST 2008


On Mon, May 12, 2008 at 10:50:03AM +0100, Federico Calboli wrote:
> 
> On 12 May 2008, at 01:05, Andrew Robinson wrote:
> 
> >On Mon, May 12, 2008 at 10:34:40AM +1200, Rolf Turner wrote:
> >>
> >>On 12/05/2008, at 9:45 AM, Andrew Robinson wrote:
> >>
> >>>On Sun, May 11, 2008 at 07:52:50PM +0100, Federico Calboli wrote:
> >>>>
> >>>>The main point of my question is, having a 3 way anova (or  
> >>>>ancova, if
> >>>>you prefer), with *no* nesting, 2 fixed effects and 1 random  
> >>>>effect,
> >>>>why is it so boneheaded difficult to specify a bog standard fully
> >>>>crossed model? I'm not talking about some rarified esoteric model
> >>>>here, we're talking about stuff tought in a first year Biology  
> >>>>Stats
> >>>>course here[1].
> >>>
> >>>That may be so, but I've never needed to use one.
> >>
> >>	So what?  This is still a standard, common, garden-variety
> >>	model that you will encounter in exercises in many (if not
> >>	all!) textbooks on experimental design and anova.
> >
> >To reply in similar vein, so what?  Why should R-core or the R
> >community feel it necessary to reproduce every textbook example?  How
> >many times have *you* used such a model in real statistical work,
> >Rolf?
> 
> There is a very important reason why R (or any other stats package)  
> should *easily* face the challenge of bog standard models: because it  
> is a *tool* for an end (i.e. the analysis of data to figure out what  
> the heck they tell us) rather than a end in itself.

But a tool that mostly (entirely?) appears in textbooks.  
 
> Bog standard models are *likely* to be used over and over again  
> because they are *bog standard*, and they became such by being used  
> *lots*.

Well.  I have documentation relevant to nlme that goes back about 10
years.  I don't know when it was first added to S-plus, but I assume
that it was about then.  Now, do you think that if the thing that you
want to do was really bog standard, that noone would have raised a
fuss or solved it within 10 years?
 
> If someone with a relatively easy model cannot use R for his job s/he  
> will use something else, and the R community will *not* increase in  
> numbers. Since R is a *community driven project*, you do the math on  
> what that would mean in the long run.

Fewer pestering questions?  ;)

Andrew

-- 
Andrew Robinson  
Department of Mathematics and Statistics            Tel: +61-3-8344-6410
University of Melbourne, VIC 3010 Australia         Fax: +61-3-8344-4599
http://www.ms.unimelb.edu.au/~andrewpr
http://blogs.mbs.edu/fishing-in-the-bay/



More information about the R-help mailing list