re: st: 3 Level Nested Anova with equal sample sizes?

[David Airey <david.airey@vanderbilt.edu>]

I'm not sure I quite understand but I have the feeling I disagree with 
what you are saying. The reason I don't understand is that it sounds 
like if you buy the machine you get specific software. Therefore you 
are not able to separate the effects of machine and software. I would 
think you can not run one company's software on another company's 
machine. But in your schematic of the design, you have software crossed 
with machine (all softwares used on all machines), not nested, which 
suggests you could run company A's software on company B's machine? If 
one is not able to uncouple software from machine, which seems likely, 
then really you just have a simple one-way ANOVA for any given measure, 
where the replicate in a cell is subject? That is, you are comparing 
four ways of getting you measure, person, machinesoftware1, 
machinesoftware2, and machinesoftware3. Since you have 30 measures you 
might consider a oneway MANOVA. Apologies if I have completely lost the 
thread of what you intended to convey.

-Dave

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/