[R] problem with se.contrast()

[Christoph Buser]

Dear Jamie

I already thought that your data structure could be more
complicated than in the example. 
I would be careful anywhere. Since there is a difference in the
results of se.contrasts() in R-devel and the results from lme
(and the with lme consistent results of the aggregated data)
in this nice balanced design, I am suspicious, especially if you
real design is more complicated.
Even if you get no error message for your data, I'd calculate
the analysis using for example lme for a second time to control
the results. 
I wish you all the best for your analysis.

Christoph Buser

--------------------------------------------------------------
Christoph Buser <buser at stat.math.ethz.ch>
Seminar fuer Statistik, LEO C11
ETH (Federal Inst. Technology)	8092 Zurich	 SWITZERLAND
phone: x-41-1-632-5414		fax: 632-1228
http://stat.ethz.ch/~buser/
--------------------------------------------------------------


Jamie Jarabek writes:
 > Christoph,
 > 
 > Thank you for your advice. My actual design is indeed more complicated than what 
 > I have indicated here. I was just using this as a toy example illustrate my 
 > particular problem. As suggested by Prof. Ripley I will download R-devel and see 
 > if the fixes included within alleviate my problems.
 > 
 > Jamie Jarabek
 >