Skip to content

how to summarize results from studies?

6 messages · Michael Dewey, Frederik Feys, Marc Schwartz

#
Hello everyone

I have some studies with results from the same outcome scale. I want to merge them into 1 summarised estimated result and its standard deviation. How do I do that in R?

Thank you very much for your help!
Frederik
#
Hi,

It sounds like you will want to engage in a meta-analysis.

There is a CRAN task view here:

  https://cran.r-project.org/web/views/MetaAnalysis.html

that would be relevant in pointing you to tools in R that can support that approach.

That being said, the details of specific methodologies and conceptual assistance would be beyond the scope of this list. You should consider consulting a local statistician for assistance with that, if needed.

Regards,

Marc Schwartz
#
Dear Frederik

There is also a mailing list dedicated to meta-analysis in R

https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis//

Michael
On 01/07/2020 16:40, Marc Schwartz via R-help wrote:

  
    
  
#
Thank you Marc!

I ended up using metafor library:

res_UK <- escalc(mi=data_UK$GAD.7_mean, sdi=data_UK$weight_pred, ni=data_UK$GAD.7_mean_N, measure = "MN?)

rma(yi, vi, data=res_UK, method="REML?)
#
Thank you Michael!
#
Michael,

Thanks for the reminder on that.

Frederik, sometimes, with an emphasis on sometimes, the r-sig-* lists are willing to go beyond narrow R programming assistance, and offer domain specific conceptual assistance, which would otherwise be off-topic for r-help.

You might look through the archives of that list to get a sense for the subject matter that has been covered there:

  https://stat.ethz.ch/pipermail/r-sig-meta-analysis/ <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/>

I might also mention that a Google search using "meta analysis in R", comes up with a number of relevant hits, including various tutorials.

Regards,

Marc