Re: st:meta-analysis

[Tom Trikalinos <ttrikalin@mac.com>]

Generally speaking, if you want to check whether any factor(s) X1 
(X2...) affect(s) Theta (ie the Effect Size), a meta-regression 
framework is probably warranted:
Try using _metareg_ and check the documentation.

. findit metareg

. metareg Theta X1 X2,  wsse(seTheta)

Really advanced analysis would be based on Bayesian approaches and is 
relatively easily implemented in WinBUGS (free specialist software, try 
googling it).  Lots of papers in Stat Med on Bayesian meta-analysis.

Hope this helps,

Tom



On Sep 3, 2004, at 11:11 AM, cui liwei wrote:


> thanks Tom, but I still want to know if the sample size will affect 
> the results.
> you see, the "meta" doesn't take into account the "n".
>
> input n theta se_theta
> 18 4.37 1.0740
> 18 3.13 1.5602
> 20 3.51 1.8622
> 21 3.34 1.9122
> 18 4.25 .60029
> end
>
> meta theta se_theta
>
>
>
>
>
> > From: Tom Trikalinos <ttrikalin@mac.com>
> > Reply-To: statalist@hsphsun2.harvard.edu
> > To: statalist@hsphsun2.harvard.edu
> > Subject: Re: st:meta-analysis
> > Date: Tue, 31 Aug 2004 11:59:48 +0300
> >
> > I suppose you know what you're doing;
> > . findit metan
> > OR
> > .findit meta
> >
> > yields useful routines. Use Coef values as the effect size (Theta) 
> > and the SE thereof as SE_theta.
> >
> > Good luck
> >
> > Tom
> >
> >
> > On Aug 31, 2004, at 11:44 AM, cui liwei wrote:
> >
> >
> > > Dear all,
> > >
> > > Here are the results of several regress,how to perform meta-analysis.
> > > thanks!
> > >
> > > Obs	Coef.	Std.Err.	t	P>|t|
> > > 18	4.37	1.0740	4.07	0.001
> > > 18	3.13	1.5602	2.01	0.062
> > > .
> > > .
> > > 18	4.25	.60029	7.08	0.000
> > >
> > > Cui
> > >
> > > _________________________________________________________________
> > > ~~{SkA*;z5DEsSQ=xPP=;Aw#,GkJ9SC~~} MSN Messenger:  
> > > http://messenger.msn.com/cn
> > > *
> > > *   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/
> _________________________________________________________________
> ~{SkA*;z5DEsSQ=xPP=;Aw#,GkJ9SC~} MSN Messenger:  http://messenger.msn.com/cn
> *
> *   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/