Hi,
what about this
// suppose you have 4 studies
clear
input study cases total
1 20 1000
2 40 5000
3 30 1500
4 25 3300
end
gen p = .
gen se = .
// get proportions and std errors
forv i =1(1)4 {
cii total[`i'] cases[`i']
qui replace p = r(mean) in `i'
qui replace se = r(se) in `i'
}
// get the inverse variance-weighted proportion
// use the official Stata -vwls- command
gen cons =1
vwls p cons, sd(se)
// use the user written -metan- command
// fixed-effects meta-analysis
metan p se, nograph fixed
// random-effects meta-analysis
metan p se, nograph random
Nicola
b. water wrote:
> Thanks Svend,
>
> Not quite sure that is legit. Could be biased by study(ies) from
> higher/lower risk populations, which would skew their rates. I think
> there is no Stata user solution. At the moment, I am thinking of using
> 'the inverse variance-weighted' method to pool. Vaguely I think there
> was a criticism about this method but would have to do for now.
>
> Other advice, help and opinion are still welcome.
>
> Thanks once again.
>
> Regards,
> bw
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today - it's
> FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> *
> * 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/
