1
$\begingroup$

This question: Difference between Cohen's d and Hedges' g for effect size metrics shows there are at least two different interpretations each of both Cohen’s D and Hedge’s G, one of them in common (the one with weighted pooled standard deviation calculated using weights proportional to n-1, where n is the sample size). This confusion (at least in the moment when I am writing) may also be seen in this Wikipedia page: https://en.wikipedia.org/wiki/Effect_size , where this common formula is shown for both statistics (in the case of Hegde’s G, a star is added to $s$ in the standard deviation, but then the formulas for $s$ and $s^*$ are the same). It seems to me the situation is even more complicated than what seen in Stack Exhange’s questions. In fact, I see several websites using an unweighted version of $s$ (so that it is simply the squared root of the mean between $s_1^2$ and $s_2^2$): https://www.statisticshowto.com/cohens-d/ ; https://www.socscistatistics.com/effectsize/default3.aspx ; https://www.statology.org/interpret-cohens-d/ ; https://www.itl.nist.gov/div898/software/dataplot/refman2/auxillar/hedgeg.htm . I see here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3840331/ that this is defined as “common language effect size” (on the StackExchange search engine, by typing "common language effect size" I find 14 questions on it). Given the confusion arising from these contrasting definitions, wouldn’t it be possible to reach a consensus to solve this issue?

$\endgroup$

1 Answer 1

1
$\begingroup$

I find this article a good starting point. The author uses subscripts to distinguish the different version of Cohen's d.

https://www.frontiersin.org/articles/10.3389/fpsyg.2013.00863/full

$\endgroup$
1
  • $\begingroup$ Thank you. You just posted another link where finding the paper I also posted in my question: "Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs", calling the unweighted version of Cohen's D as "common language effect size". $\endgroup$ Commented Jun 27, 2022 at 7:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.