Quote:
Originally Posted by Ersby  Cohen's h concerns itself with describing ratios, as far as I can tell. In terms of a ganzfeld experiment, that would mean an experiment with 10 trials and 3 hits would have the same effect size h as an experiment with 100 trials and 30 hits. That's my understanding, at least. |
Yes, that's my understanding too. A Cohen's h score in this case would take the proportion of hits relative to chance for observed and expected so the value would be the same in these two situations. I think it would actually be about 0.43 if my calculations are right.
What I can't figure out is what it means to transform the aspirin data into Cohen's h, with regards to making an effect size comparison between psi and the aspirin study.