[R-meta] Compiling different design in the same met-analysis
||@t@ @end|ng |rom dewey@myzen@co@uk
Mon May 3 12:40:59 CEST 2021
On 03/05/2021 08:13, Gladys Barragan-Jason wrote:
> Dear all,
> I am doing a meta-analysis on experimental data on the impact on
> several behavioral interventions including both *pre-post studies*
> (measurement of a trait before and after the intervention) *and control
> vs. treatment studies *(measurement after a control intervention vs.
> measurement after the intervention).
So the first design just has one group which is measured twice and the
second design has two parallel groups which are each measured once?
> I am interested in the effect size of each type of intervention on a
> specific trait that is common to all studies.
> I would like to know whether it is correct to compile all studies from
> one type of intervention (whatever the type of comparisons: pre-pot and
> control vs. treatment group together) with pre- and control being as
> control and post and treatment being as treatment.
I do not think that is going to work well. Although the two designs
could be considered as estimating the same effect their variability
should be very different since in design 1 you have reduced the effect
of individual variability. If you do include both designs in one
analysis I would have thought it mandatory to include a moderator with
two levels for design.
> I have a second question about how to thank the
> r-sig-meta-analysis using r-project.org
> <mailto:r-sig-meta-analysis using r-project.org> list for all their help
> (acknowledgements in a paper).
I would have thought the URL
was better. How to do this will also depend on custom and practice in
your scientific discipline, some journals in mine require people who are
acknowledged to give their consent.
> Thanks a lot for your help!
> Have a great day,
> Gladys Barragan-Jason, PhD. ____ Website
> Station d'Ecologie Théorique et Expérimentale (SETE)____
> CNRS de Moulis____
> Virus-free. www.avg.com
> R-sig-meta-analysis mailing list
> R-sig-meta-analysis using r-project.org
More information about the R-sig-meta-analysis