>As such, it's likely that this study will find, entirely correctly, that recipients are happier, more productive, etc. I wonder what the right way is to measu

>As such, it's likely that this study will find, entirely correctly, that recipients are happier, more productive, etc. I wonder what the right way is to measure happiness for a group. Say this test makes 20% of participants much happier, 70% about the same, and 10% overdose on heroine? And what metric can you use to quantify happiness on an individual level?