Page 1 of 1

Major-Meta Analysis of Proportions

Posted: Fri Apr 26, 2024 4:11 am
by njain@iiphg.org
I am writing to discuss a challenge we encountered when we ran the meta-analysis in the JAMOVI cloud.

We wanted to run a proportions meta-analysis to calculate the pooled prevalence. However, after several attempts, we could not perform the test.

A message “the condition has length >1” is displayed.

Please let us know if you have any thoughts or recommendations on how we can proceed. Your support is invaluable.

Re: Major-Meta Analysis of Proportions

Posted: Sat Apr 27, 2024 12:38 pm
by MAgojam
njain@iiphg.org wrote: Fri Apr 26, 2024 4:11 am Please let us know if you have any thoughts or recommendations on how we can proceed. Your support is invaluable.
Hey Nidhi,
Jonathon called my attention to your post, because W. Kyle Hamilton author and creator of the MAJOR module we haven't seen here for a long time.
If MAJOR presents bugs in use, I could intervene on the module code to fix them, as I have already done in the past.
To find out more, I would need your jamovi file with a set of data (even false ones) in which the selected analysis leads to the error and to know which version of jamovi you are using and for which operating system.

Consider whether my request is feasible.

Cheers,
Maurizio Agosti
https://www.jamovi.org/about.html

Re: Major-Meta Analysis of Proportions

Posted: Sun Apr 28, 2024 7:13 am
by angela
I have encountered the same problem. Here is my set of data. I am running the dichotomous model.

Re: Major-Meta Analysis of Proportions

Posted: Mon Apr 29, 2024 4:31 am
by njain@iiphg.org

Re: Major-Meta Analysis of Proportions

Posted: Mon Apr 29, 2024 12:18 pm
by MAgojam
Thanks @angela for sharing your Epi.omv file.

Thanks @nidhi for the link, but it is not shareable and accessible (only you can access it).
If you want you can save and compress it on your PC and attach it as a reply here on the forum.


However, I have identified the problems that are occurring with MAJOR, which under the hood has a very old version of the R metafor package and the current jamovi updates require heavy intervention on some initial modules, such as MAJOR.
I'll have to spend some time "regenerating it to new life", perhaps even listing those updates.
Unfortunately this won't happen anytime soon, because there are things going on that take up all the time I can spend here.
For this reason I feel the duty to suggest that you also take a look at the Meta-Analysis possible with the esci module by Robert J Calin-Jageman present in the library.

Cheers,
Maurizio Agosti
https://www.jamovi.org/about.html