st: R: Anova repeated measures

["Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>]

Instead of a repeated ANOVA, and warning about the missing values (which may
raise substantial issues in his analysis), Peter may want to perform an OLS
on vas as dependent variable with clustered standard error on projektnum,
being this one the principal sample unit.

----------------- begin code ------------------
g gruppe_bis=0 if   gruppe=="Parenteral morphine"
replace gruppe_bis=1 if gruppe=="XXX with NaCl"
replace gruppe_bis=2 if   gruppe=="XXX with Naropine"
reg vas gruppe_bis, cluster(projektnum)
----------------- end code ------------------

HTH and Kind Regards,
Carlo
-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Peter Kraglund
Jacobsen
Inviato: mercoledì 21 aprile 2010 22.24
A: statalist@hsphsun2.harvard.edu
Oggetto: st: Anova repeated measures

projektnum	t	gruppe	vas
93	1	Parenteral morphine	8
93	2	Parenteral morphine	.
93	3	Parenteral morphine	9
93	4	Parenteral morphine	9
93	5	Parenteral morphine	8
95	1	XXX with NaCl	5
95	2	XXX with NaCl	.
95	3	XXX with NaCl	3
95	4	XXX with NaCl	2
95	5	XXX with NaCl	.
101	1	XXX with Naropine	3
101	2	XXX with Naropine	.
101	3	XXX with Naropine	5
101	4	XXX with Naropine	5
101	5	XXX with Naropine	.

In this example, three persons (projektnum) in three groups (gruppe)
are asked about pain score (vas) at five different times (t).
How do I show if there is a difference in VAS between the three groups?
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/