Re: st: Repeated anove problems

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

.

You need to explicitly put the id term in the model, and you need to  
specify the repeated factors, not the id, in the repeated option.  
There is an FAQ on a two factor within subjects design like yours.  
Your two repeated factors are site and phase.

http://www.stata.com/support/faqs/stat/anova2.html#half713

anova value id site / id*site phase / id*phase site*phase,  
repeated(site phase)
A couple other FAQs tell you about how to follow up a significant  
omnibus F tests with post hoc tests, because the omnibus F tests  
provided by ANOVA are only half the story, when you have more than 2  
levels in a factor.


You can also try this model with manova or xtmixed.



-Dave


On Mar 23, 2008, at 11:57 AM, S�ren O'Neill wrote:

> My data represents pressure-pain-threshold measurements  
> (variable=value) from
> 13 individuals (variable=id), from 8 sites (in each individual)
> (variable=site) at 3 differents times (variable=phase) - before,  
> during and
> after intervention.
>
> I'm trying to do a 2-way, repeated-measures anove, to see if time  
> and/or site
> is significant.
>
> I run:
> anove value site phase, repeated(id)
>
> I get the following error: term not in model
>
> I've made sure data are de-stringed - so thats not the problem.
>
> Any suggestions?
> --
> Med venlig hilsen  /  Kind regards
> S�ren O'Neill
>
> vCard : http://www.oneill.dk/Soren_ONeill.vcf
>
> *
> *   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/

*
*   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/