[Feature Request]: Meta-analysis: Add Orwin’s Fail-Safe N

[giligg]

• Enhancement: adding Orwin’s Fail-Safe N for the file-drawer analysis.
• Purpose: Currently included is the Rosenthal Fail-Safe N analysis which has several shortcoming and is less reported in meta-analses papers. Orwin’s method allows for more flexibility in the analysis by not assuming the missing effect size are equal to 0, so it can be specified missing effect sizes in any value. Also we can determine what effect size other than 0 we want the reduce the overall effect to.

[tomtomme]

Still valid for 0.19 beta

[FBartos]

I will do a larger revamp of the meta-analysis module this year and keep this in mind