Hello everyone
Just downloaded Jamovi days ago and I want to thank all who made this software possible.
I have a question about Subgroup analysis after getting the original forest plot of meta-analysis; can we do it in Jamovi? and How (if possible)?
Thanks a lot!
Subgroup analysis after Meta analysis
Re: Subgroup analysis after Meta analysis
My understanding is that you can't do this with the MAJOR package yet and it is so sad... It also really sucks that you can't unlog the forest plots. It also sucks that online for metafor R there is only an RR or OR example to copy Cochrane's vastly superior forest plot design and not for mean differences.
https://www.metafor-project.org/doku.ph ... lot_revman
https://www.metafor-project.org/doku.ph ... lot_revman
Re: Subgroup analysis after Meta analysis
So yeah you are kinda screwed there as am I. The only option is to use Cochrane's RevMan, however you are locked to use Dersimonian-Laird estimator (DL) which, as opposed to Restricted ML (REML) is proven to suck with high heterogeneity data and will mistakenly produce significant results when there shouldn't be.
So here we are in 2022 and there is still no easy to use an adequate open-source solution for meta-analysis.
So here we are in 2022 and there is still no easy to use an adequate open-source solution for meta-analysis.
Re: Subgroup analysis after Meta analysis
Thank you for your feedback..really appreciated.
Re: Subgroup analysis after Meta analysis
Just found this about JASP here:https://github.com/jasp-stats/jasp-issues/issues/373, I am not 100% sure if this applies to Jamovi too, but I think will apply.
"Sub-analyses are possible with the case filter feature, and running a meta-analysis on subsets of cases. But combining them in forest plots as in the example is not implemented. If I remember correctly I decided to not add that mainly because it would require an additional user interface, and you can simply do it with sub-grouping your data and running a meta analysis."
"Sub-analyses are possible with the case filter feature, and running a meta-analysis on subsets of cases. But combining them in forest plots as in the example is not implemented. If I remember correctly I decided to not add that mainly because it would require an additional user interface, and you can simply do it with sub-grouping your data and running a meta analysis."
Re: Subgroup analysis after Meta analysis
yes, if all you're trying to do is analyse a subset of your data, you can just use filters.
jonathon
jonathon
Re: Subgroup analysis after Meta analysis
I've also been using Jamovi for a while, and this hasn't been possible. Is the above link your project?apavlo wrote:My understanding is that you can't do this with the MAJOR package yet and it is so sad... It also really sucks that you can't unlog the forest plots. It also sucks that online for metafor R there is only an RR or OR example to copy Cochrane's vastly superior forest plot design and not for mean differences.
https://www.metafor-project.org/doku.ph ... lot_revman