st: question about SE when running _metan_ w/ random effects

["Erick Turner" <turnere@ohsu.edu>]

Dear Stata users of meta-analytic programs,

I am using Stata, specifically  -metan-, to run a random effect meta- 
analysis on continuous data. I have calculate Hedges' g. I also used  
the sample sizes (with g) to calculate the SE around each g, which I  
gather is more for a fixed effect model.

If needed, I can also calculate the SE for the difference between the  
means (but not individually for each mean).

What I have done so far is to run -metan-, plugging in the values for  
Hedges' g and SE (from the sample sizes), then ask for random effects.

I used the dialog box approach (the "Effect/SE" button under the  
"Main" tab), but here's how the command line came out:

 metan g se, by(drug) random

Am I doing this correctly? In particular, is it OK to use the SE  
derived from the sample sizes (and g)? If not, what would you suggest?

Thanks very much.

Erick




*
*   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/