[R-meta] Help computing effect size of two-choice test (binary response)

Michael Dewey ||@t@ @end|ng |rom dewey@myzen@co@uk
Wed Nov 3 10:04:30 CET 2021

Dear Nathalia

The other two effect sizes you mention are comparative ones (SMD and OR) 
so proportions will not be comparable to them.

Why would you want to convert them to another metric? Proportions have a 
straightforward intuitive interpretation, everyone knows what a half 
means, but who knows what g=0.5 means without some thought? If at the 
end of your analysis you can say that predators prefer untreated prey 
0.75 of the time then does that not answer your scientific question?


On 03/11/2021 01:31, Nathalia Ximenes Gonçalves wrote:
> Dear Wolfgang,
> Many thanks for your answer! I don't have any experience with proportions
> in meta-analysis. Are they comparable to standardized means and odds ratio?
> Is it possible to convert proportions to hedges's g?
> Thanks,
> Nathalia
> Em ter., 2 de nov. de 2021 às 03:32, Viechtbauer, Wolfgang (SP) <
> wolfgang.viechtbauer using maastrichtuniversity.nl> escreveu:
>> Dear Nathalia,
>> Based on this description, it seems to me that all that you can estimate
>> from these studies is the proportion (or some transformation thereof) of
>> predators that choose prey with chemical defenses. The choice (prey with
>> chemical defense or prey without chemical defense) is an outcome, not a
>> 'treament'.
>> Best,
>> Wolfgang
>>> -----Original Message-----
>>> From: R-sig-meta-analysis [mailto:
>> r-sig-meta-analysis-bounces using r-project.org] On
>>> Behalf Of Nathalia Ximenes Gonçalves
>>> Sent: Monday, 01 November, 2021 19:18
>>> To: r-sig-meta-analysis using r-project.org
>>> Subject: [R-meta] Help computing effect size of two-choice test (binary
>> response)
>>> Hello everyone,
>>> I'm conducting a meta-analysis about chemical defenses effectiveness. I'm
>>> facing a problem with a set of studies and I was wondering if someone
>> could
>>> help me. In some studies, the experiment was conducted in a two-choice
>>> test, in which the predator had to choose a prey with chemical defense or
>> a
>>> prey without chemical defense. For instance, if the experiment had 20
>>> predators, each one had to choose between two types of prey, let's suppose
>>> that 16 predators had choosen prey without chemical defenses and 4
>>> predators had choosen prey with chemical defenses. Therefore, I would
>> have:
>>> 'treatment = 4', 'control = 16', and a total sample size of 20. I cannot
>>> use any odds ratio, risk ratio, etc metrics, because I don't have a 2x2
>>> table, and all the results would be equal to zero if I consider a sample
>>> size of 20 for treatment and control separately. Would anyone have any
>>> recommendations for calculating the effect size for this type of study?
>>> Thanks,
>>> Nathalia
> 	[[alternative HTML version deleted]]
> _______________________________________________
> R-sig-meta-analysis mailing list
> R-sig-meta-analysis using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis


More information about the R-sig-meta-analysis mailing list