[R-meta] effect size similarities
That is just a simple z-test of a single binomial proportion. This cannot be turned into a standardized mean difference (Cohen's d / Hedges' g value) or vice-versa. Best, Wolfgang
-----Original Message----- From: Patrizio E Tressoldi [mailto:patrizio.tressoldi at unipd.it] Sent: Tuesday, 22 June, 2021 16:01 To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis at r-project.org Subject: Re: [R-meta] effect size similarities Il 22/06/2021 15:54, Viechtbauer, Wolfgang (SP) ha scritto:
Dear Patrizio, I don't know what exactly you mean by 'binomial z value', but Hedges' g is not a
'z score'. If that were the case, then g values would have a variance of 1, which is clearly not true.
Here you can see the formula: https://www.mathcelebrity.com/binomial.php?n=10&p=0.4&k=3&t=&pl=Normal+Approximati on <https://www.mathcelebrity.com/binomial.php?n=10&p=0.4&k=3&t=&pl=Normal+Approximat ion> Given you said they are different, how they must be transformed for their comparison? Thanks Patrizio