No I would never make that leap. I believe social sciences in particular have suffered from over simplifying human behaviour to reduce it down to something that

No I would never make that leap. I believe social sciences in particular have suffered from over simplifying human behaviour to reduce it down to something that appears testable and that has resulted in silly predictions/claims. But medical trials is a whole other story. Here you can very clearly construct a double blind to test the efficacy of a medicine.