Meta-Analysis

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
lia
Posts: 2
Joined: Fri Jul 17, 2020 2:22 pm

Meta-Analysis

Post by lia »

Hi,
I'm currently conduction a Meta-Analysis and am a bit confused by how to interpret the result. The formula I usually use is I2 = 100% * (Q – df)/Q, but when calculating it myself, using the provided Q from the heterogeneity plot, I only get the same result when using the maximum likelihood and Fisher´s r to z.

I want to do a Hunter-Schmidt analysis though and use the raw correlation and was wondering why the I2 value differs so much and if the interpretation of the Q and I2 are still the same.

I also wasn't able to use the corrected correlation (its says I need the gsl package, but I couldn't find one in the library) and couldn't figure out how to solve that problem.

I attached two screenshots.


thank you,
Lia
Attachments
I2= 44% instead of almost 83%
I2= 44% instead of almost 83%
Screen Shot 2020-07-17 at 4.41.20 pm.png (28.76 KiB) Viewed 1744 times
bias corrected
bias corrected
Screen Shot 2020-07-17 at 4.40.23 pm.png (37.15 KiB) Viewed 1744 times
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Meta-Analysis

Post by jonathon »

hi lia,

i've forwarded your query to kyle hamilton, and hopefully he'll come here and respond.

cheers

jonathon
lia
Posts: 2
Joined: Fri Jul 17, 2020 2:22 pm

Re: Meta-Analysis

Post by lia »

thank you Jonathon!
Post Reply