This doesn't feel like a very scientific approach. Looking at national satisfaction and then national advertising spend...seems like there would be huge scope f

This doesn't feel like a very scientific approach. Looking at national satisfaction and then national advertising spend...seems like there would be huge scope for bias regardless of how many controls you put in. In particular, something like individual wellbeing is just that...individual. Really very tricky (and as always you see the researcher give out ludicrously unjustified advice to individuals...if you go to a psychologist and tell them you are unhappy, they don't ask how many adverts you saw recently).