Where is the effect size and what does it mean.

From All Medications Are Insignificant In The Eyes Of God And Traditional Effect Size Criteria by Scott Alexander in Astral Codex Ten.  

A long and detailed essay, rewarding of the time to read.

SSRI antidepressants like Prozac were first developed in the 1980s and 1990s. The first round of studies, sponsored by pharmaceutical companies, showed they worked great! But skeptics found substantial bias in these early trials; several later analyses that corrected for this all found effect sizes (compared to placebo) of only 0.30.

Is an effect size of 0.30 good or bad? The usual answer is “bad”. The UK’s National Institute for Clinical Excellence used to say that treatments were only “clinically relevant” if they had an effect size of 0.50 or more. The US FDA apparently has a rule of thumb that any effect size below 0.50 is “small”.

Others are even stricter. Leucht et al investigate when doctors subjectively feel like their patients have gotten better, and find that even effect size 0.50 correlates to doctors saying they see little or no change. Based on this research, Irving Kirsch, author of some of the earliest credible antidepressant effect size estimates, argues that “[the] thresholds suggested by NICE were not empirically based and are presumably too small”, and says that “minimal improvement” should be defined as an effect size of 0.875 or more. No antidepressant consistently attains this. 

He then goes on to point out this absence of effect size is in itself misleading.   Basically, there are multiple measures which are narrowly useful but you have to stitch together multiple, often contradictory of one another, measures before even beginning to understand something.  

Effect size is indeed important.  But it is not the only measure of importance.