This study accounts for that in all the experiments described. They send financial problems at random to people. That means the genes of the individuals will be

This study accounts for that in all the experiments described. They send financial problems at random to people. That means the genes of the individuals will be random. The sugar farmers in India also present a good random experiment because they are not picking people based on whether they are rich or poor already, but based on how things turn out from one rather random event: the harvest. Saying categorically: "Any social study that doesn't control for genes is a worthless study. Do a twin study or don't even bother." Just shows ignorance on the matter. 1) Genetics wont affect a lot of studies. 2) Twin studies are not the gold standard you seem believe it is. Adopted children in twin studies are usually not adopted into a random environment, since the requirements are high. When the environment is stable and has little variation, then most of the variation you measure will be heritable, thus falsely setting heritability very high.