Hi Daniela Vogel,

I have a personal believe that any standardized effect size (like Cohen's d, Hedges' g and so on) are useless if you have everything measured in the same way.

Those metrics do not tell you about clinical relevance (common mistake) and do not add anything in terms of comprehension beyond a point estimate and its confidence intervals. The only reason to use standardized effect sizes is when you have different measurements/scales and want to combine then.

I go beyond that: when I see standardized effect sizes in those scenarios, the risk for p-hacking for me rises significantly!!!!!

Thanks,

LH