I'm somewhat doubtful of the design as it's explained in this blogpost, seems like there will be lots of treatment externalities. I'd really recommend anyone wh

I'm somewhat doubtful of the design as it's explained in this blogpost, seems like there will be lots of treatment externalities. I'd really recommend anyone who's interested read http://cega.berkeley.edu/assets/cega_research_projects/1/Identifying-Impacts-on-Education-and-Health-in-the-Presence-of-Treatment-Externalities.pdf http://cega.berkeley.edu/assets/cega_research_projects/1/Ide... for an explanation of why a totally randomised control trial is the wrong choice for things like this.