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
Meta-Analysis
Meta-Analysis
- Attachments
-
- I2= 44% instead of almost 83%
- Screen Shot 2020-07-17 at 4.41.20 pm.png (28.76 KiB) Viewed 1774 times
-
- bias corrected
- Screen Shot 2020-07-17 at 4.40.23 pm.png (37.15 KiB) Viewed 1774 times
Re: Meta-Analysis
hi lia,
i've forwarded your query to kyle hamilton, and hopefully he'll come here and respond.
cheers
jonathon
i've forwarded your query to kyle hamilton, and hopefully he'll come here and respond.
cheers
jonathon