So in the code, perhaps just replace the D with your a priori effect size, as you allude at the bottom of the blog. ie, the starting mean and the SD may come from the studies data, but the "D" would certainly not.
For me, when I read a study that reports no effect, then I may sceptically do a power analysis on that data, I would be doing exactly this: (as Dr Anon says) Calculating power given the SD of the study, but an effect size of something I would have a priori considered biologically important. Yes both, rather belatedly(!!) I think there is a misunderstanding here, due to the language of stats! On reading your blog I realised I would be mis-using the term "post-hoc power". It was also discussed in Schimmack (2012) as a problem in the averaging of observed power as an estimate of true power and was the reason why the replicability index uses the median to estimate true (median) power of a set of studies.
50 was discussed in Yuan and Maxwell (2005) ģ. The skewed distribution of observed power for true power unequal. It would suggest that researchers have only a 60% chance to get a significant effect in a replication study and should increase power to make a replication effort worthwhile.Ģ. Even if median power is 60% and only 60% of results are significant, post-hoc power is informative. I would not trust an article that reports 10 significant results when the median power is 60%. However, post-hoc power provides valuable information for a set of independent statistical tests. I would like to add that post-hoc power for a single statistical test is useless.
0 kommentar(er)
