Re: st: meta-analysis

[Rebecca Boehm <boehm@musc.edu>]

As I understand it, as long as you have the risk estimates and standard 
errors, meta is suitable. It will also give the forest plot. Funnel plots 
to assess publication bias, with either Begg's or Egger's style graphs, can 
be gotten with metabias. And finally, if you have heterogeneity to explore, 
metareg is the way to go.

I just completed my own meta-analysis using these programs. I hope that 
they are similarly useful to you.

Becky Boehm
Ph.D track, Biostatistics
Department of Biometry and Epidemiology
Medical University of South Carolina

--On Wednesday, April 23, 2003 3:51 PM +0100 Patrick Sturgis 
<P.Sturgis@soc.surrey.ac.uk> wrote:


> I have 16 test-retest studies from which have been calculated the
> before-after difference in Cronbach's Alpha for multi-item scales. The
> standard error of the difference in Alpha has been estimated using the
> bootstrap. I now wish to conduct a meta-analysis of these studies. Can
> anyone suggest a suitable Stata user-written program for doing this? Would
> Meta be appropriate? Thanks in advance,
>
> Patrick
>
> Dr Patrick Sturgis
> Department of Sociology
> University of Surrey
> Guildford
> Surrey GU2 7XH
> United Kingdom
> tel: +44 (0)1483 686 973
> fax: +44(0)1483 689551
>
>
> *
> *   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/