[R-meta] Multilevel meta-analysis using z-scores

Viechtbauer, Wolfgang (SP) wolfg@ng@viechtb@uer @ending from m@@@trichtuniver@ity@nl
Tue May 22 17:32:21 CEST 2018

Hi Gergő,

Could you explain in more detail how you computed/extracted these 'z-scores/t-values'? This is not a standard outcome measure as used in meta-analyses, so without further details about their computation, it's hard for me to give any advice.


-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces at r-project.org] On Behalf Of BARANYI Gergö
Sent: Tuesday, 22 May, 2018 17:18
To: r-sig-meta-analysis at r-project.org
Subject: [R-meta] Multilevel meta-analysis using z-scores

Dear All,
We are working on a review about how different environmental exposures (e.g. green space) influence particular psychiatric disorders, by only including longitudinal studies. As the effect parameters, statistical models and study types (e.g. RCT, cohort studies, quasi-experimental studies) of the included studies are very different, we found calculating z-scores/t-values the best way the analyse the data. Furthermore, the data has also a multilevel structure, where z-values with different follow-up times are nested within the same sample. 
Unfortunately, I have never used the metaphor package before, but after reading the R help file, I understood, that multilevel mixed-effects meta-analysis (rma.mv) would best fit to our project as we have extracted also several possible moderator variables. However, I did not find the way in "escalc" calculating sampling weights from the study sample size and from z-score or impute them directly in the multilevel model.
As I am still learning R, I would be really thankful for any suggestion or advices how to solve this problem.
Thank you very much for your help!

More information about the R-sig-meta-analysis mailing list